INDEX

Explanations

programming syntax and technical constructs related to code definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 Bulk

-0.17

éĻº

-0.15

 bulk

-0.15

ëŁ¬ìĬ¤

-0.15

íķŃ

-0.14

coni

-0.14

[string

-0.14

 bÃ³ng

-0.14

unittest

-0.14

 camp

-0.14

POSITIVE LOGITS

 Pont

0.19

Pont

0.17

imos

0.17

ability

0.17

org

0.16

innamon

0.15

#line

0.15

idlo

0.15

com

0.15

ype

0.14

Activations Density 0.007%

programming syntax and technical constructs related to code definitions

No Comments

No Known Activations