INDEX

Explanations

references to educational events and opportunities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

æ½

-0.15

 MÃľ

-0.14

 Hakk

-0.14

noop

-0.14

chu

-0.14

igkeit

-0.14

querque

-0.13

 Ashton

-0.13

inez

-0.13

POSITIVE LOGITS

 prevailing

0.19

 nett

0.18

GI

0.18

 Enjoy

0.16

nett

0.16

 kindly

0.16

.POS

0.15

 Flat

0.15

 Counter

0.15

 quotation

0.15

Activations Density 0.032%

references to educational events and opportunities

No Comments

No Known Activations