INDEX

Explanations

terms related to kinetic energy and various classifications of "K" words, often in the context of scientific or technical subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

-0.16

 conoc

-0.15

levant

-0.14

gether

-0.14

acco

-0.14

igne

-0.14

apo

-0.14

shire

-0.14

omm

-0.13

kker

-0.13

POSITIVE LOGITS

nap

0.23

rypton

0.20

 vá»±c

0.17

elsey

0.17

etchup

0.17

ãĤĤãģĹãĤĮãģªãģĦ

0.17

fold

0.17

ety

0.17

itchens

0.16

patrick

0.16

Activations Density 0.697%

terms related to kinetic energy and various classifications of "K" words, often in the context of scientific or technical subjects

No Comments

No Known Activations