INDEX

Explanations

technical terms related to measurement and precision

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-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_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

 normal

-0.15

ongan

-0.15

int

-0.14

atrix

-0.14

ÙĪØªÛĮ

-0.14

itori

-0.14

 proportion

-0.13

optimize

-0.13

AccessException

-0.13

 easiest

-0.13

POSITIVE LOGITS

 accuracy

0.65

 precision

0.56

 Accuracy

0.54

accuracy

0.51

 accur

0.50

 Precision

0.45

Accuracy

0.45

 accurate

0.44

Precision

0.42

precision

0.42

Activations Density 0.134%

technical terms related to measurement and precision

No Comments

No Known Activations