Update README.md

README.md

@@ -7,19 +7,20 @@ base_model:
 - bartowski/google_gemma-3-1b-it-GGUF
 ---
 
-This is a "self" merge of https://huggingface.co/google/gemma-3-1b-it-qat-q4_0-gguf and https://huggingface.co/bartowski/google_gemma-3-1b-it-GGUF.
+This is a "self" merge of https://huggingface.co/google/gemma-3-1b-it-qat-q4_0-gguff.
 
-The official QAT weights released by google use fp16 (instead of Q6_K) for the embeddings table, which makes this model take a significant extra amount of memory (and storage) compared to what Q4_0 quants are supposed to take. Instead of quantizing the table myself, I extracted it from Bartowski's quantized models, because those were already calibrated with imatrix, which should squeeze some extra performance out of it.
+The official QAT weights released by google use fp16 (instead of Q6_K) for the embeddings table, which makes this model take a significant extra amount of memory (and storage) compared to what Q4_0 quants are supposed to take. 
+~~Instead of quantizing the table myself, I extracted it from Bartowski's quantized models, because those were already calibrated with imatrix, which should squeeze some extra performance out of it.~~
+Requantizing with llama.cpp fixes that and gives better result than the other thing.
 
 Here are some perplexity measurements:
 
 | Model | File size ↓ | PPL (wiki.text.raw) ↓ |
 | --- | --- | --- |
-| [This model](https://huggingface.co/stduhpf/google-gemma-3-1b-it-qat-q4_0-gguf-small/blob/main/gemma-3-1b-it-q4_0_s.gguf) | 720 MB | 28.2603 +/- 0.26947 |
+| [This model](https://huggingface.co/stduhpf/google-gemma-3-1b-it-qat-q4_0-gguf-small/blob/main/gemma-3-1b-it-q4_0_s.gguf) | 720 MB | 28.0468 +/- 0.26681 |
+| [This model (older version)](https://huggingface.co/stduhpf/google-gemma-3-1b-it-qat-q4_0-gguf-small/blob/f325927302d106ad204c0b6a8f09f216a0447519/gemma-3-1b-it-q4_0_s.gguf) | 720 MB | 28.2603 +/- 0.26947 |
 | [Q4_0 (bartowski)](https://huggingface.co/bartowski/google_gemma-3-1b-it-GGUF/blob/main/google_gemma-3-1b-it-Q4_0.gguf) | 722 MB | 34.4906 +/- 0.34539 |
 | [QAT Q4_0 (google)](https://huggingface.co/google/gemma-3-1b-it-qat-q4_0-gguf/blob/main/gemma-3-1b-it-q4_0.gguf) | 1 GB | 28.0400 +/- 0.26669 |
 | [BF16 (upscaled to f32 for faster inference)](https://huggingface.co/google/gemma-3-1b-it) | 2 GB | 29.1129 +/- 0.28170 |
 
-Note that this model ends up smaller than the Q4_0 from Bartowski. This is because llama.cpp sets some tensors to Q4_1 when quantizing models to Q4_0 with imatrix, but this is a static quant.
-The perplexity scores are barely within margin of error between this model and the original QAT, it seems like the embedding table starts making a difference at this small size, though the trade off is probably still worth it. 
-Interestingly perplexity seems sigificantly better for both QAT models compared to BF16.
+Note that this model ends up smaller than the Q4_0 from Bartowski. This is because llama.cpp sets some tensors to Q4_1 when quantizing models to Q4_0 with imatrix, but this is a static quant.