diff --git a/presentation/claude-code-best-practice/index.html b/presentation/claude-code-best-practice/index.html index 3ce3142..eddd5d4 100644 --- a/presentation/claude-code-best-practice/index.html +++ b/presentation/claude-code-best-practice/index.html @@ -497,25 +497,13 @@
-
- - -

Tokens, not words

- - -
- Screenshot of the OpenAI tokenizer showing the sentence about BPE split into 32 tokens across 105 characters, with tabs for GPT-5.x, GPT-4, and GPT-3 tokenizers. -
- 105 characters → 32 tokens. Rule of thumb: ~4 chars per token in English.
- Each model generation uses a different tokenizer — same text, different token count, different cost. -
-
- -
+
+ Screenshot of the OpenAI tokenizer showing the sentence about BPE split into 32 tokens across 105 characters, with tabs for GPT-5.x, GPT-4, and GPT-3 tokenizers. +
@@ -556,13 +544,12 @@
-
-

💬 Models are stateless

+

💬 Models are stateless

-
+
@@ -605,10 +592,9 @@
-

Every turn is a fresh API call.

-

Memory only exists if the harness replays the transcript.

+

Every turn is a fresh API call.

+

Memory only exists if the harness replays the transcript.

-
@@ -617,11 +603,8 @@
-
-
🦜
-

“Stochastic Parrots”

-

Stochastic means random/probabilistic — models don’t know the answer, they sample from a probability distribution.

-
+

🦜 “Stochastic Parrots”

+

Stochastic means random/probabilistic — models don’t know the answer, they sample from a probability distribution.

@@ -687,10 +670,8 @@
-
-

Even temperature = 0 isn’t deterministic.

-

You set it to zero. You expect the same answer every time. You’re wrong.

-
+

🌡️ Even temperature = 0 isn’t deterministic.

+

You set it to zero. You expect the same answer every time. You’re wrong.