INDEX

Explanations

mentions of memes and discussions around them

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

con

-0.18

-0.17

ise

-0.16

olie

-0.16

uer

-0.15

inf

-0.14

 segment

-0.14

olumn

-0.14

 gains

-0.14

 mole

-0.14

POSITIVE LOGITS

chner

0.17

;element

0.16

htar

0.15

Mov

0.15

akens

0.15

ADDE

0.15

ÇĲ

0.15

PerPixel

0.14

ilk

0.14

amma

0.14

Activations Density 0.007%

mentions of memes and discussions around them

No Comments

No Known Activations