INDEX

Explanations

performance metrics in racing contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

éľŀ

-0.16

TL

-0.15

 Rope

-0.15

amba

-0.15

riel

-0.14

æĪ²

-0.14

UIL

-0.14

 hinge

-0.14

elsea

-0.13

 foss

-0.13

POSITIVE LOGITS

 Course

0.16

ovol

0.15

 finish

0.15

preg

0.15

estead

0.15

 ascending

0.15

 course

0.15

à¤°à¤£

0.14

udy

0.14

 finishing

0.14

Activations Density 0.045%

performance metrics in racing contexts

No Known Activations