summaryrefslogtreecommitdiff
path: root/config/eval_gpt2.py
diff options
context:
space:
mode:
authorVoid Agent <void@jayrup.hermes>2026-07-29 17:47:48 +0100
committerVoid Agent <void@jayrup.hermes>2026-07-29 17:47:48 +0100
commitdb4ec5bb3839bc5cc50d82e427848595d14b3070 (patch)
treed2d9fb11e11392077a79e6a10c38d44d1d9f5a38 /config/eval_gpt2.py
parent66f99ee30087a5f28ad852e581a0334c7f556091 (diff)
Restructure: nanoGPT at root, custom code in src/
- Move model.py, train.py, configurator.py to root for nanoGPT compatibility - data/ and config/ directories at root with Shakespeare dataset prep scripts - src/jlens.py updated to import model from project root - Cleaned up stale src/config/ and duplicate src/ files - Fixed .gitignore: exclude out-shakespeare-char/ instead of raw data dirs
Diffstat (limited to 'config/eval_gpt2.py')
-rw-r--r--config/eval_gpt2.py8
1 files changed, 8 insertions, 0 deletions
diff --git a/config/eval_gpt2.py b/config/eval_gpt2.py
new file mode 100644
index 0000000..53978cb
--- /dev/null
+++ b/config/eval_gpt2.py
@@ -0,0 +1,8 @@
+# evaluate the base gpt2
+# n_layer=12, n_head=12, n_embd=768
+# 124M parameters
+batch_size = 8
+eval_iters = 500 # use more iterations to get good estimate
+eval_only = True
+wandb_log = False
+init_from = 'gpt2'