From 564e986496219fd5a2abd1dfa1221d65e0a9735a Mon Sep 17 00:00:00 2001 From: casinca <47400729+casinca@users.noreply.github.com> Date: Sun, 22 Jun 2025 19:15:01 +0200 Subject: [PATCH] fix issue #664 - inverted token and pos emb layers (#665) * fix inverted token and pos layers * remove redundant code --------- Co-authored-by: rasbt --- .../exercise-solutions.ipynb | 20 ++++++------------- 1 file changed, 6 insertions(+), 14 deletions(-) diff --git a/ch02/01_main-chapter-code/exercise-solutions.ipynb b/ch02/01_main-chapter-code/exercise-solutions.ipynb index 52d8f90..77b3774 100644 --- a/ch02/01_main-chapter-code/exercise-solutions.ipynb +++ b/ch02/01_main-chapter-code/exercise-solutions.ipynb @@ -46,8 +46,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "torch version: 2.4.0\n", - "tiktoken version: 0.7.0\n" + "torch version: 2.6.0\n", + "tiktoken version: 0.9.0\n" ] } ], @@ -327,21 +327,13 @@ " raw_text = f.read()\n", "\n", "tokenizer = tiktoken.get_encoding(\"gpt2\")\n", - "encoded_text = tokenizer.encode(raw_text)\n", - "\n", - "vocab_size = 50257\n", - "output_dim = 256\n", - "max_len = 4\n", - "context_length = max_len\n", - "\n", - "token_embedding_layer = torch.nn.Embedding(context_length, output_dim)\n", - "pos_embedding_layer = torch.nn.Embedding(vocab_size, output_dim)" + "encoded_text = tokenizer.encode(raw_text)" ] }, { "cell_type": "code", "execution_count": 13, - "id": "0128eefa-d7c8-4f76-9851-566dfa7c3745", + "id": "15c184fe-5553-4df2-a77f-7504901b6709", "metadata": {}, "outputs": [ { @@ -371,7 +363,7 @@ { "cell_type": "code", "execution_count": 14, - "id": "ff5c1e90-c6de-4a87-adf6-7e19f603291c", + "id": "739990b2-ce4c-4d17-88e3-547c8c312019", "metadata": {}, "outputs": [ { @@ -415,7 +407,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.11.4" + "version": "3.10.16" } }, "nbformat": 4,