yujunjun/LLMs-from-scratch
1
0
Fork 0
mirror of https://github.com/rasbt/LLMs-from-scratch.git synced 2025-08-28 18:40:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

change defaults to 0 temp

Browse Source
This commit is contained in:
rasbt 2024-05-19 09:04:49 -05:00
parent 59f5ed8d68
commit bc5cbbf1bd
4 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
ch05/01_main-chapter-code/ch05.ipynb
View File

@ -1852,7 +1852,7 @@
"metadata": {}, "metadata": {},
"outputs": [], "outputs": [],
"source": [ "source": [
"def generate(model, idx, max_new_tokens, context_size, temperature=1.0, top_k=None, eos_id=None):\n", "def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None):\n",
"\n", "\n",
" # For-loop is the same as before: Get logits, and only focus on last time step\n", " # For-loop is the same as before: Get logits, and only focus on last time step\n",
" for _ in range(max_new_tokens):\n", " for _ in range(max_new_tokens):\n",
@ -2442,7 +2442,7 @@
"name": "python", "name": "python",
"nbconvert_exporter": "python", "nbconvert_exporter": "python",
"pygments_lexer": "ipython3", "pygments_lexer": "ipython3",
"version": "3.11.4" "version": "3.10.6"
} }
}, },
"nbformat": 4, "nbformat": 4,

2
ch05/01_main-chapter-code/gpt_generate.py
View File

@ -215,7 +215,7 @@ def load_weights_into_gpt(gpt, params):
gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"]) gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"])
def generate(model, idx, max_new_tokens, context_size, temperature, top_k=None, eos_id=None): def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None):
# For-loop is the same as before: Get logits, and only focus on last time step # For-loop is the same as before: Get logits, and only focus on last time step
for _ in range(max_new_tokens): for _ in range(max_new_tokens):

2
ch05/02_alternative_weight_loading/previous_chapters.py
View File

@ -254,7 +254,7 @@ def token_ids_to_text(token_ids, tokenizer):
return tokenizer.decode(flat.tolist()) return tokenizer.decode(flat.tolist())
def generate(model, idx, max_new_tokens, context_size, temperature=1.0, top_k=None, eos_id=None): def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None):
# For-loop is the same as before: Get logits, and only focus on last time step # For-loop is the same as before: Get logits, and only focus on last time step
for _ in range(max_new_tokens): for _ in range(max_new_tokens):

2
ch06/02_bonus_additional-experiments/previous_chapters.py
View File

@ -316,7 +316,7 @@ def load_weights_into_gpt(gpt, params):
gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"]) gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"])
def generate(model, idx, max_new_tokens, context_size, temperature=1.0, top_k=None, eos_id=None): def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None):
# For-loop is the same as before: Get logits, and only focus on last time step # For-loop is the same as before: Get logits, and only focus on last time step
for _ in range(max_new_tokens): for _ in range(max_new_tokens):
idx_cond = idx[:, -context_size:] idx_cond = idx[:, -context_size:]