Merge 69b7cc9ac5 into bc1fca39f3

Remove redundant uv venv from shell scripts.

nanochat/gpt.py

@@ -8,7 +8,7 @@ Notable features:
 - norm after token embedding
 - no learnable params in rmsnorm
 - no bias in linear layers
-- Multi-Query Attention (MQA) support for more efficient inference
+- Group-Query Attention (GQA) support for more efficient inference
 """
 
 import math
@@ -29,7 +29,7 @@ class GPTConfig:
     vocab_size: int = 50304
     n_layer: int = 12
     n_head: int = 6 # number of query heads
-    n_kv_head: int = 6 # number of key/value heads (MQA)
+    n_kv_head: int = 6 # number of key/value heads (GQA)
     n_embd: int = 768
 
 

run1000.sh

@@ -9,7 +9,6 @@ export OMP_NUM_THREADS=1
 export NANOCHAT_BASE_DIR="$HOME/.cache/nanochat"
 mkdir -p $NANOCHAT_BASE_DIR
 command -v uv &> /dev/null || curl -LsSf https://astral.sh/uv/install.sh | sh
-[ -d ".venv" ] || uv venv
 uv sync --extra gpu
 source .venv/bin/activate
 if [ -z "$WANDB_RUN" ]; then

speedrun.sh

@@ -20,9 +20,7 @@ mkdir -p $NANOCHAT_BASE_DIR
 
 # install uv (if not already installed)
 command -v uv &> /dev/null || curl -LsSf https://astral.sh/uv/install.sh | sh
-# create a .venv local virtual environment (if it doesn't exist)
-[ -d ".venv" ] || uv venv
-# install the repo dependencies
+# create a .venv local virtual environment (if it doesn't exist) and install the repo dependencies
 uv sync --extra gpu
 # activate venv so that `python` uses the project's venv instead of system python
 source .venv/bin/activate