INDEX

Explanations

punctuation and emotionally charged phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

.bunifuFlatButton

-0.15

erken

-0.15

braco

-0.15

 respectively

-0.14

orthand

-0.14

amba

-0.14

 Leban

-0.13

ãĥĥãĤ·ãĥ¥

-0.13

Trap

-0.13

gre

-0.13

POSITIVE LOGITS

ecz

0.16

sole

0.15

 Civilization

0.14

 migrationBuilder

0.14

 recent

0.13

resden

0.13

hattan

0.13

 Ø¹ÙĤØ¯

0.13

uds

0.13

ļĮ

0.13

Activations Density 0.031%

punctuation and emotionally charged phrases

No Comments

No Known Activations