INDEX

Explanations

references to magazine issues and publication details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ught

-0.15

 Separate

-0.14

afil

-0.14

acr

-0.14

ovel

-0.14

à¸¤

-0.13

Ess

-0.13

adioButton

-0.13

vod

-0.13

 Stock

-0.13

POSITIVE LOGITS

 Magazine

0.17

 issue

0.17

 magazine

0.17

egin

0.16

issue

0.16

mag

0.15

azine

0.15

 magazines

0.14

TION

0.14

ccione

0.14

Activations Density 0.042%

references to magazine issues and publication details

No Comments

No Known Activations