INDEX

Explanations

concepts related to recovery and returning to a previous state or condition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

agy

-0.17

inke

-0.15

Realm

-0.14

oni

-0.14

åĩºãģĹãģŁ

-0.14

ÃŃÅ¾

-0.14

commend

-0.14

loy

-0.14

print

-0.14

 print

-0.13

POSITIVE LOGITS

odos

0.17

ple

0.16

iyat

0.16

keh

0.15

strcasecmp

0.15

igans

0.14

icl

0.14

itive

0.14

 Conway

0.14

berapa

0.14

Activations Density 0.393%

concepts related to recovery and returning to a previous state or condition

No Comments

No Known Activations