INDEX

Explanations

programming and technical language constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

ÙĨØ¯Ø§ÙĨ

-0.16

ê¶ģ

-0.16

irie

-0.16

uct

-0.16

roleum

-0.15

Ð¸Ð½Ð¾Ð²

-0.14

taire

-0.14

ampus

-0.14

dens

-0.14

UCT

-0.14

POSITIVE LOGITS

uby

0.17

hil

0.15

uzzi

0.14

uppe

0.14

opis

0.14

ä»ĺãģĳ

0.14

 Dough

0.14

ReadWrite

0.14

agens

0.14

Vij

0.14

Activations Density 0.001%

programming and technical language constructs

No Comments

No Known Activations