INDEX

Explanations

terms related to loss, damage, and modification in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

thouse

-0.17

activex

-0.15

 KÃ¼r

-0.14

ail

-0.14

RESS

-0.14

Ax

-0.14

utura

-0.13

 ãĥĻ

-0.13

ress

-0.13

lando

-0.13

POSITIVE LOGITS

 Histogram

0.14

zens

0.14

.hist

0.14

 éĦ

0.14

excel

0.14

 Lomb

0.13

Ð²Ð°Ð»Ð°

0.13

ÙģØª

0.13

Ð°Ð½Ðº

0.13

 Diaz

0.13

Activations Density 0.357%

terms related to loss, damage, and modification in various contexts

No Comments

No Known Activations