INDEX

Explanations

mentions of official statements or reports from authoritative sources

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

illion

-0.15

ertools

-0.15

pec

-0.15

 Butler

-0.14

 Carroll

-0.14

imizer

-0.13

dlg

-0.13

ossible

-0.13

xz

-0.13

æ·

-0.13

POSITIVE LOGITS

hait

0.19

 OMIT

0.17

icerca

0.15

PostalCodes

0.15

 ÙħØ´Ø§Ùĩ

0.15

.ActionListener

0.14

ugen

0.14

olu

0.14

 Kosten

0.14

ehr

0.13

Activations Density 0.115%

mentions of official statements or reports from authoritative sources

No Comments

No Known Activations