INDEX

Explanations

references to weapons and tools

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

Ð¾ÑģÐ¿

-0.16

enor

-0.16

Ø§ÙĨÙĩ

-0.16

-Sah

-0.15

ëĭµ

-0.15

CFG

-0.14

ehr

-0.14

BarItem

-0.14

agos

-0.14

specifier

-0.14

POSITIVE LOGITS

 holder

0.25

 usage

0.23

 wield

0.22

 holders

0.22

-shaped

0.22

 wielding

0.22

 handle

0.22

 Holder

0.21

 bearer

0.21

-holder

0.21

Activations Density 0.207%

references to weapons and tools

No Comments

No Known Activations

references to weapons and tools

No Comments

No Known Activations