INDEX

Explanations

terms related to architecture and prefabrication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

á»ģn

-0.15

 keyed

-0.14

 Continent

-0.14

resco

-0.14

 ÙĨÚ©

-0.14

braco

-0.14

arga

-0.14

 turf

-0.13

urname

-0.13

/all

-0.13

POSITIVE LOGITS

Ã©gor

0.19

meer

0.16

uzu

0.15

rops

0.14

racÃŃ

0.14

semi

0.14

mlink

0.13

inkel

0.13

lesi

0.13

linky

0.13

Activations Density 0.012%

terms related to architecture and prefabrication

No Comments

No Known Activations