INDEX

Explanations

elements related to rejection, disappointment, and failure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

.pkg

-0.15

inou

-0.15

UBLE

-0.15

kip

-0.15

ZF

-0.14

allee

-0.14

ndon

-0.14

levator

-0.14

fy

-0.13

#ga

-0.13

POSITIVE LOGITS

ola

0.17

æ®¿

0.16

æİī

0.16

Ä°K

0.15

olas

0.15

oola

0.15

 unless

0.15

ikk

0.14

remium

0.14

ibal

0.14

Activations Density 0.368%

elements related to rejection, disappointment, and failure

No Comments

No Known Activations