INDEX

Explanations

mathematical notations and structures related to proofs and theorems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 nightly

-0.16

Ã¢h

-0.14

reib

-0.14

figcaption

-0.13

æĺ¯åľ¨

-0.12

olist

-0.12

</

-0.12

POSITIVE LOGITS

intro

0.15

PERTIES

0.14

_intro

0.14

_codegen

0.13

¼åĲĪ

0.13

 intro

0.13

+-

0.13

 ZeroConstructor

0.12

,www

0.12

Intro

0.12

Activations Density 0.129%

mathematical notations and structures related to proofs and theorems

No Comments

No Known Activations