INDEX

Explanations

negative descriptors or terms related to limitation and prohibition

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

aled

-0.15

 IBOutlet

-0.15

AVE

-0.15

inst

-0.15

immers

-0.15

oller

-0.14

alon

-0.13

arton

-0.13

à¸´à¸Ĺ

-0.13

elo

-0.13

POSITIVE LOGITS

deo

0.15

acios

0.15

Ð¾ÑģÐ¿

0.14

ssid

0.14

 createState

0.14

Ã©ri

0.14

ÑĥÐ´Ð°

0.14

umnos

0.13

 Listings

0.13

 Ð´Ñĥ

0.13

Activations Density 0.029%

negative descriptors or terms related to limitation and prohibition

No Comments

No Known Activations