INDEX

Explanations

references to specific educational institutions or programs, particularly those associated with "Upper" and "Lower" designations

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

 bottom

-0.20

top

-0.19

.top

-0.18

 bottoms

-0.17

ONSE

-0.16

top

-0.16

bottom

-0.16

Top

-0.15

-top

-0.15

unn

-0.15

POSITIVE LOGITS

elon

0.19

hone

0.17

amak

0.16

hle

0.15

ibbon

0.15

eso

0.15

horn

0.14

_cast

0.14

SCORE

0.14

 Exact

0.14

Activations Density 0.021%

references to specific educational institutions or programs, particularly those associated with "Upper" and "Lower" designations

No Comments

No Known Activations