INDEX

Explanations

terms associated with academic research and publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

earer

-0.15

SWG

-0.15

laÄį

-0.14

cke

-0.13

klass

-0.13

.GetKey

-0.13

.paper

-0.13

rex

-0.13

ght

-0.13

_CLASS

-0.13

POSITIVE LOGITS

quine

0.13

pek

0.13

ekl

0.13

å¿ĥéĩĮ

0.13

unden

0.13

\b

0.13

FP

0.13

 hostile

0.13

bolt

0.13

eteor

0.13

Activations Density 0.002%

terms associated with academic research and publications

No Comments

No Known Activations