INDEX

Explanations

references to coding and programming terminology within the text

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

">//

-0.15

 Ø§ÙĦØ¹Ùħ

-0.14

aida

-0.14

$MESS

-0.14

arov

-0.14

 taraf

-0.14

íķľêµŃ

-0.14

á»¥c

-0.13

vey

-0.13

ÃŃch

-0.13

POSITIVE LOGITS

ETING

0.16

abee

0.15

.ci

0.15

 corr

0.15

Ø§Ø¬Ùĩ

0.14

_EPS

0.14

 Herb

0.14

 Timber

0.14

angers

0.14

bis

0.13

Activations Density 0.001%

references to coding and programming terminology within the text

No Comments

No Known Activations