INDEX

Explanations

references to television series and their episodic structure

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

New Auto-Interp

Configuration

fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_mlp_out

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

ica

-0.15

ICA

-0.15

stell

-0.15

zzo

-0.15

IC

-0.15

ua

-0.15

 RENDER

-0.14

 -------------------------------------------------------------------------↵

-0.14

inson

-0.14

ocode

-0.14

POSITIVE LOGITS

acin

0.17

pap

0.17

IFS

0.15

Î±Î½Î´

0.14

ÑıÑī

0.14

 enthusiasts

0.14

StandardItem

0.14

_native

0.14

ãĥ¼ãĥĨãĤ£

0.14

à¹Ĥà¸Ĺ

0.14

Activations Density 0.006%

references to television series and their episodic structure

No Comments

No Known Activations