Llama-3-Giraffe-70B

Abacus.AI presents our longer-necked variant of Llama 3 70B!

This model has an effective context length of approximately 128k.

We have currently trained on ~1B tokens. This is an initial release and we are hoping to improve the heatmap below further as we continue training.

Training Methodology

The methodology for training uses PoSE and dynamic-NTK interpolation.

NTK-scaling

The scale factor for NTK is 4. Note that we also tried theta-scaling but this did not work as well as NTK scaling in our experiments.

PoSE

We utilise Positional Skip-wise Training (PoSE) with the following parameters:

• Number of Chunks: 5
• Max position ID: 32768

Data

We use on average ~8K long samples from RedPajama.

Hardware

We train on 8xH100 GPUs with Deepspeed Zero Stage 3.

Evaluation Methodology

We use the EasyContext implementation of Needle-in-a-Haystack to evaluate Llama-3-Giraffe-70B.

We evaluate with the following parameters:

• Min context length: 2000
• Max context length: 128000
• Context interval: 4000
• Depth interval: 0.1
• Num samples: 2
• Rnd number digits: 7
• Haystack dir: PaulGrahamEssays