INDEX

Explanations

references to original or custom-made items and components

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

ille

-0.17

isko

-0.17

anner

-0.17

anker

-0.16

 purs

-0.15

uent

-0.15

sonian

-0.15

anto

-0.15

 ingr

-0.14

POSITIVE LOGITS

own

0.18

 proprietary

0.16

<fieldset

0.16

 reusable

0.15

ADIUS

0.15

 prÃ³p

0.14

own

0.14

 ÑģÐ¾Ð±ÑģÑĤÐ²ÐµÐ½

0.14

 escorte

0.14

 unique

0.14

Activations Density 0.204%

references to original or custom-made items and components

No Comments

No Known Activations