INDEX

Explanations

references to research citations and academic studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

AGON

-0.17

Bry

-0.15

 hypothetical

-0.14

leigh

-0.14

asil

-0.14

irthday

-0.13

anguard

-0.13

anner

-0.13

Virgin

-0.13

agon

-0.13

POSITIVE LOGITS

å¿µ

0.14

.MSG

0.14

tent

0.14

0.13

dba

0.13

.Circle

0.13

 Buffer

0.13

plat

0.13

çĶº

0.12

upy

0.12

Activations Density 0.014%

references to research citations and academic studies

No Comments

No Known Activations