INDEX

Explanations

terms related to quarrels or conflicts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

-0.17

edly

-0.16

-0.15

ebra

-0.15

hev

-0.15

cents

-0.15

Ori

-0.14

 Ã¸

-0.14

_ABI

-0.14

POSITIVE LOGITS

term

0.26

antine

0.25

rel

0.24

rels

0.24

ries

0.24

rell

0.21

tern

0.21

coop

0.20

REL

0.19

rel

0.19

Activations Density 0.004%

terms related to quarrels or conflicts

No Comments

No Known Activations