INDEX

Explanations

references to energy consumption and efficiency metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

adelphia

-0.15

rej

-0.15

 Dysfunction

-0.14

 Frequ

-0.14

deen

-0.14

icontrol

-0.14

onica

-0.14

ZA

-0.14

eparator

-0.13

 Dtype

-0.13

POSITIVE LOGITS

 consumption

0.43

 usage

0.36

 requirements

0.36

 requirement

0.33

consum

0.31

usage

0.31

 Consumption

0.31

Consum

0.31

requirements

0.30

 consum

0.30

Activations Density 0.164%

references to energy consumption and efficiency metrics

No Comments

No Known Activations

references to energy consumption and efficiency metrics

No Comments

No Known Activations