INDEX

Explanations

structured information and references related to educational resources and platforms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

elters

-0.16

estone

-0.14

ela

-0.14

obox

-0.14

ikel

-0.14

iap

-0.13

 Licht

-0.13

elia

-0.13

edia

-0.13

tember

-0.13

POSITIVE LOGITS

izard

0.16

unden

0.14

 ÙĩÙħÚĨÙĨÛĮÙĨ

0.14

ĨĴ

0.14

 Also

0.14

lund

0.14

ROKE

0.14

also

0.14

ndata

0.13

 especialmente

0.13

Activations Density 0.320%

structured information and references related to educational resources and platforms

No Comments

No Known Activations