INDEX

Explanations

instances of error handling in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

olv

-0.15

Ð¾Ð»Ð¾Ð¼

-0.15

 máº¡ng

-0.14

Ã©is

-0.14

Ð±Ð¾Ð¼

-0.14

ãģ£ãģį

-0.14

oj

-0.14

ìłĿ

-0.14

inia

-0.14

.coin

-0.13

POSITIVE LOGITS

agenda

0.15

 Corner

0.14

rob

0.14

spl

0.14

aternion

0.13

ughter

0.13

ÙĴÙĩ

0.13

Cab

0.13

ToString

0.13

Activations Density 0.007%

instances of error handling in code

No Comments

No Known Activations