ChiDotPhi

Mostly Theoretical Physics stuff

Weekly AI Paper - 09/05/25 - 2-Simplex attention, Mixture of Recursion, Supress wait tokens
Weekly AI Paper - 09/05/25 - 2-Simplex attention, Mixture of Recursion, Supress wait tokens
Weekly AI paper - 8/30/25 - Falcon-H1, CoT Mirage, GSPO
Weekly AI paper - 8/30/25 - Falcon-H1, CoT Mirage, GSPO
Jax tutorial 4: JIT
Jax tutorial 4: JIT
Jax tutorial 3: PMAP and VMAP
Jax tutorial 3: PMAP and VMAP
Jax tutorial 2: Grads
Jax tutorial 2: Grads
Jax tutorial 1: Arrays
Jax tutorial 1: Arrays
Weekly AI paper - 4/7/25 - LLAMA4, Multi token attention
Weekly AI paper - 4/7/25 - LLAMA4, Multi token attention
Weekly AI paper review - 3/28/25 - Large Memory module, LongRope2, MOBA block attention
Weekly AI paper review - 3/28/25 - Large Memory module, LongRope2, MOBA block attention
Weekly AI paper review - 2/14/25 - S1 Test time scaling, SMOLLM2
Weekly AI paper review - 2/14/25 - S1 Test time scaling, SMOLLM2
Weekly AI paper review - 2/7/25 - RL generalizes, Titans: Learning to Memorize at test time.
Weekly AI paper review - 2/7/25 - RL generalizes, Titans: Learning to Memorize at test time.
Weekly AI paper review - 1/23/25 - Transformers^2, Tensor Product Attention, Deepseek-R1
Weekly AI paper review - 1/23/25 - Transformers^2, Tensor Product Attention, Deepseek-R1
VSCode environment and extensions for Python Development
VSCode environment and extensions for Python Development
Weekly AI paper overview - 12/23/2024 - Deepseekv3, Memory layers
Weekly AI paper overview - 12/23/2024 - Deepseekv3, Memory layers
6. Final Words
6. Final Words
5. Training the MLLM
5. Training the MLLM
Weekly AI paper overview - 12/23/2024 - ModernBERT, No More Adam
Weekly AI paper overview - 12/23/2024 - ModernBERT, No More Adam
Weekly AI paper overview - 12/19/24 - Latent Space Reasoning , Byte Latent Transformer
Weekly AI paper overview - 12/19/24 - Latent Space Reasoning , Byte Latent Transformer
5. Dataloader for MLLM
5. Dataloader for MLLM
4. Writing MLLM from scratch
4. Writing MLLM from scratch
Torchtune: A new finetuning library for LLMs
Torchtune: A new finetuning library for LLMs
3. Vision base - Clip-14
3. Vision base - Clip-14
Weekly AI paper overview- 10/29/24 - Pixtral, Differential Transformers
Weekly AI paper overview- 10/29/24 - Pixtral, Differential Transformers
Weekly AI paper overview- 10/1/24 - LLAMA3.2, Self correction using RL
Weekly AI paper overview- 10/1/24 - LLAMA3.2, Self correction using RL
Part 4 : Trainer
Part 4 : Trainer
Part 3: Training Script
Part 3: Training Script
Part 1 : OLMO Paper
Part 1 : OLMO Paper
Part 2 : The model definition
Part 2 : The model definition
Weekly AI paper overview - 7/25/2024
Weekly AI paper overview - 7/25/2024
Weekly AI paper overview- 7/7/24
Weekly AI paper overview- 7/7/24
Weekly AI paper overview- 6/18/24
Weekly AI paper overview- 6/18/24