INDEX

Explanations

references to conferences and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ern

-0.18

ened

-0.16

red

-0.16

 liá»ĩu

-0.16

-0.15

ypad

-0.15

ombo

-0.15

nes

-0.15

erna

-0.15

fully

-0.14

POSITIVE LOGITS

ional

0.17

orio

0.16

ee

0.16

ationToken

0.15

ofs

0.15

yntax

0.15

.ibm

0.14

ESS

0.14

iversit

0.14

cko

0.14

Activations Density 0.046%

references to conferences and events

No Comments

No Known Activations