INDEX

Explanations

technical terms related to military weaponry and systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

enthal

-0.15

zte

-0.14

ibus

-0.14

oro

-0.14

andr

-0.14

 Alman

-0.13

reeze

-0.13

ATAB

-0.13

byname

-0.13

POSITIVE LOGITS

 targets

0.16

æ²¢

0.15

kov

0.15

/goto

0.14

 Shel

0.14

 versatile

0.14

-target

0.14

 targeting

0.14

edic

0.14

 target

0.14

Activations Density 0.076%

technical terms related to military weaponry and systems

No Comments

No Known Activations