INDEX

Explanations

terms related to computing instructions and user support

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ĵåĲį

-0.17

_Tis

-0.17

-inst

-0.15

bs

-0.14

imp

-0.14

mousemove

-0.14

ourcem

-0.14

imon

-0.14

ermal

-0.13

POSITIVE LOGITS

ptune

0.16

æľ¬

0.15

 Throne

0.14

¤íĶĦ

0.14

haar

0.14

æ¬¢

0.13

itori

0.13

itung

0.13

ateria

0.13

 cÃ¢

0.13

Activations Density 1.980%

terms related to computing instructions and user support

No Comments

No Known Activations