add deepseek and gemma results

1 files changed, 0 insertions, 15 deletions

diff --git a/results/classifier/deepseek-2-tmp/reasoning/graphic/696 b/results/classifier/deepseek-2-tmp/reasoning/graphic/696
deleted file mode 100644
index a7525fae..00000000
--- a/results/classifier/deepseek-2-tmp/reasoning/graphic/696
+++ /dev/null
@@ -1,15 +0,0 @@
-<think>
-Okay, so I need to classify this bug report into one of the given categories for QEMU. Let me read through the problem description carefully.
-
-The user is reporting that the EDID settings aren't being reflected in the window size when added via the command line. They're seeing the default resolution (640x480) instead of the overridden 1740x720. From their analysis, it seems that the issue lies in how the EDID is handled between different parts of QEMU.
-
-Looking at the code they mentioned, ui/console.c initializes the window size with 640x480. Then, in hw/display/virtio-gpu-base.c, there's a function called virtio_gpu_ui_info where the override isn't taking effect—it's still using the default value from console.c.
-
-The user also noted that the EDID flag is always true when they run a command, which contradicts the documentation. So, their main issue is about how the EDID information is being processed and applied to set the correct window size based on the command-line arguments.
-
-Now, considering the categories provided: mistranslation, assembly lowering, device, graphic, etc. Since this involves screen resolution and display settings, it's definitely related to graphics. The problem isn't with translations or performance but rather with how EDID is handled in the graphical output.
-
-Therefore, the category that best fits this bug report is 'graphic'.
-</think>
-
-graphic
\ No newline at end of file