INDEX

Explanations

components and processes related to fire and cooking

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

avier

-0.15

referer

-0.14

coop

-0.14

 Refriger

-0.14

-0.13

ardash

-0.13

_STORAGE

-0.13

 crush

-0.13

ainty

-0.13

PV

-0.13

POSITIVE LOGITS

çĤī

0.19

(çģ«

0.18

 fires

0.18

/fire

0.17

-hot

0.17

 Hearth

0.16

æ¸©åº¦

0.16

à¸¡à¸Ń

0.16

èħ¾

0.15

çĥ§

0.15

Activations Density 0.269%

components and processes related to fire and cooking

No Comments

No Known Activations

components and processes related to fire and cooking

No Comments

No Known Activations