Skip to content

Commit 564e986

Browse files
casincarasbt
andauthored
fix issue rasbt#664 - inverted token and pos emb layers (rasbt#665)
* fix inverted token and pos layers * remove redundant code --------- Co-authored-by: rasbt <[email protected]>
1 parent 0a2e8c3 commit 564e986

File tree

1 file changed

+6
-14
lines changed

1 file changed

+6
-14
lines changed

ch02/01_main-chapter-code/exercise-solutions.ipynb

Lines changed: 6 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -46,8 +46,8 @@
4646
"name": "stdout",
4747
"output_type": "stream",
4848
"text": [
49-
"torch version: 2.4.0\n",
50-
"tiktoken version: 0.7.0\n"
49+
"torch version: 2.6.0\n",
50+
"tiktoken version: 0.9.0\n"
5151
]
5252
}
5353
],
@@ -327,21 +327,13 @@
327327
" raw_text = f.read()\n",
328328
"\n",
329329
"tokenizer = tiktoken.get_encoding(\"gpt2\")\n",
330-
"encoded_text = tokenizer.encode(raw_text)\n",
331-
"\n",
332-
"vocab_size = 50257\n",
333-
"output_dim = 256\n",
334-
"max_len = 4\n",
335-
"context_length = max_len\n",
336-
"\n",
337-
"token_embedding_layer = torch.nn.Embedding(context_length, output_dim)\n",
338-
"pos_embedding_layer = torch.nn.Embedding(vocab_size, output_dim)"
330+
"encoded_text = tokenizer.encode(raw_text)"
339331
]
340332
},
341333
{
342334
"cell_type": "code",
343335
"execution_count": 13,
344-
"id": "0128eefa-d7c8-4f76-9851-566dfa7c3745",
336+
"id": "15c184fe-5553-4df2-a77f-7504901b6709",
345337
"metadata": {},
346338
"outputs": [
347339
{
@@ -371,7 +363,7 @@
371363
{
372364
"cell_type": "code",
373365
"execution_count": 14,
374-
"id": "ff5c1e90-c6de-4a87-adf6-7e19f603291c",
366+
"id": "739990b2-ce4c-4d17-88e3-547c8c312019",
375367
"metadata": {},
376368
"outputs": [
377369
{
@@ -415,7 +407,7 @@
415407
"name": "python",
416408
"nbconvert_exporter": "python",
417409
"pygments_lexer": "ipython3",
418-
"version": "3.11.4"
410+
"version": "3.10.16"
419411
}
420412
},
421413
"nbformat": 4,

0 commit comments

Comments
 (0)