INDEX

Explanations

elements related to specific venues and experiences

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

CED

-0.15

vit

-0.15

plib

-0.15

ãİ

-0.14

 OMIT

-0.14

ternet

-0.14

uish

-0.14

abit

-0.13

adolu

-0.13

flo

-0.13

POSITIVE LOGITS

iones

0.16

irm

0.15

ukan

0.14

il

0.14

ucc

0.14

è¦ª

0.14

enate

0.14

Harness

0.13

rack

0.13

issan

0.13

Activations Density 0.092%

elements related to specific venues and experiences

No Comments

No Known Activations