INDEX

Explanations

requests for assistance or help in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

orris

-0.14

ohn

-0.14

 Merry

-0.14

.pageY

-0.14

]={↵

-0.13

intree

-0.13

ÏĮÎ½

-0.13

Triple

-0.13

wap

-0.13

/cs

-0.13

POSITIVE LOGITS

 Guinness

0.14

is

0.14

AEA

0.14

 mere

0.14

Fuse

0.14

acha

0.13

 Kelley

0.13

 emerging

0.13

åīĩ

0.13

 Flag

0.13

Activations Density 0.328%

requests for assistance or help in various contexts

No Comments

No Known Activations