Re-upload MLX conversion from google/gemma-4-E2B-it@70af34e20bd4b7a91f0de6b22675850c43922a03

README.md

@@ -1,25 +1,25 @@
 ---
 library_name: mlx
-license: apache-2.0
-license_link: https://ai.google.dev/gemma/docs/gemma_4_license
-pipeline_tag: any-to-any
-base_model: google/gemma-4-e2b-it
+license: gemma
+base_model: google/gemma-4-E2B-it
 tags:
 - mlx
+- mlx-vlm
+- gemma4
+- image-text-to-text
 ---
 
 # mlx-community/gemma-4-e2b-it-8bit
 
-This model was converted to MLX format from [`google/gemma-4-e2b-it`](https://huggingface.co/google/gemma-4-e2b-it)
-using mlx-vlm version **0.4.3**.
-Refer to the [original model card](https://huggingface.co/google/gemma-4-e2b-it) for more details on the model.
+MLX conversion of [google/gemma-4-E2B-it](https://huggingface.co/google/gemma-4-E2B-it) for Apple silicon.
 
-## Use with mlx
+- Source revision: `70af34e20bd4b7a91f0de6b22675850c43922a03`
+- Variant: `8bit`
+- Converted with `mlx_vlm.convert` from the local `mlx-vlm` checkout.
 
-```bash
-pip install -U mlx-vlm
-```
+## Usage
 
 ```bash
-python -m mlx_vlm.generate --model mlx-community/gemma-4-e2b-it-8bit --max-tokens 100 --temperature 0.0 --prompt "Describe this image." --image <path_to_image>
+pip install mlx-vlm
+python -m mlx_vlm.generate --model mlx-community/gemma-4-e2b-it-8bit --prompt "Describe this image." --image path/to/image.jpg
 ```

config.json

@@ -45,6 +45,7 @@
     "audio_token_id": 258881,
     "boa_token_id": 256000,
     "boi_token_id": 255999,
+    "do_sample": true,
     "dtype": "bfloat16",
     "eoa_token_id": 258883,
     "eoa_token_index": 258883,
@@ -54,6 +55,20 @@
         106,
         50
     ],
+    "generation_config": {
+        "bos_token_id": 2,
+        "do_sample": true,
+        "eos_token_id": [
+            1,
+            106,
+            50
+        ],
+        "pad_token_id": 0,
+        "temperature": 1.0,
+        "top_k": 64,
+        "top_p": 0.95,
+        "transformers_version": "5.5.0.dev0"
+    },
     "image_token_id": 258880,
     "initializer_range": 0.02,
     "model_type": "gemma4",
@@ -67,6 +82,7 @@
         "bits": 8,
         "mode": "affine"
     },
+    "temperature": 1.0,
     "text_config": {
         "attention_bias": false,
         "attention_dropout": 0.0,
@@ -152,6 +168,8 @@
         "vocab_size_per_layer_input": 262144
     },
     "tie_word_embeddings": true,
+    "top_k": 64,
+    "top_p": 0.95,
     "transformers_version": "5.5.0.dev0",
     "video_token_id": 258884,
     "vision_config": {

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe889fb027f0b79758af4a7da6a27c6c7bc715680bbdd5af9797bd8355d86820
-size 5367135201
+oid sha256:0e82359bc2f0effdad15168d05a2ae3d46e72cb3296b5658303a2f78eee9926b
+size 5365634455

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83bb2a3420d473d416ffcb3cf9c93bacce064981fb22ea20cb6111a178d2679b
-size 532432577
+oid sha256:797c7786d137de2b77826ae838ee366eab5a9479dfca91b345f71b7e99d71cfe
+size 500946703

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
     "metadata": {
-        "total_size": 5899232198
+        "total_size": 5866263494
     },
     "weight_map": {
         "audio_tower.layers.0.feed_forward1.ffw_layer_1.input_max": "model-00001-of-00002.safetensors",
@@ -77,34 +77,34 @@
         "audio_tower.layers.1.feed_forward1.ffw_layer_2.output_min": "model-00001-of-00002.safetensors",
         "audio_tower.layers.1.feed_forward1.post_layer_norm.weight": "model-00001-of-00002.safetensors",
         "audio_tower.layers.1.feed_forward1.pre_layer_norm.weight": "model-00001-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_1.input_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_1.input_min": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_1.linear.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_1.output_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_1.output_min": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_2.input_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_2.input_min": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_2.linear.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_2.output_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.ffw_layer_2.output_min": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.post_layer_norm.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.feed_forward2.pre_layer_norm.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.conv_norm.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.depthwise_conv1d.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_end.input_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_end.input_min": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_end.linear.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_end.output_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_end.output_min": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_start.input_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_start.input_min": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_start.linear.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_start.output_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.lconv1d.linear_start.output_min": "model-00002-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_1.input_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_1.input_min": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_1.linear.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_1.output_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_1.output_min": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_2.input_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_2.input_min": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_2.linear.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_2.output_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.ffw_layer_2.output_min": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.post_layer_norm.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.feed_forward2.pre_layer_norm.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.conv_norm.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.depthwise_conv1d.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_end.input_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_end.input_min": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_end.linear.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_end.output_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_end.output_min": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_start.input_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_start.input_min": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_start.linear.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_start.output_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.lconv1d.linear_start.output_min": "model-00001-of-00002.safetensors",
         "audio_tower.layers.1.lconv1d.pre_layer_norm.weight": "model-00001-of-00002.safetensors",
-        "audio_tower.layers.1.norm_out.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.norm_post_attn.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.1.norm_pre_attn.weight": "model-00002-of-00002.safetensors",
+        "audio_tower.layers.1.norm_out.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.norm_post_attn.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.1.norm_pre_attn.weight": "model-00001-of-00002.safetensors",
         "audio_tower.layers.1.self_attn.k_proj.input_max": "model-00001-of-00002.safetensors",
         "audio_tower.layers.1.self_attn.k_proj.input_min": "model-00001-of-00002.safetensors",
         "audio_tower.layers.1.self_attn.k_proj.linear.weight": "model-00001-of-00002.safetensors",
@@ -251,18 +251,18 @@
         "audio_tower.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors",
         "audio_tower.layers.11.self_attn.v_proj.output_max": "model-00002-of-00002.safetensors",
         "audio_tower.layers.11.self_attn.v_proj.output_min": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.2.feed_forward1.ffw_layer_1.input_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.2.feed_forward1.ffw_layer_1.input_min": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.2.feed_forward1.ffw_layer_1.linear.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.2.feed_forward1.ffw_layer_1.output_max": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.2.feed_forward1.ffw_layer_1.output_min": "model-00002-of-00002.safetensors",
+        "audio_tower.layers.2.feed_forward1.ffw_layer_1.input_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.2.feed_forward1.ffw_layer_1.input_min": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.2.feed_forward1.ffw_layer_1.linear.weight": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.2.feed_forward1.ffw_layer_1.output_max": "model-00001-of-00002.safetensors",
+        "audio_tower.layers.2.feed_forward1.ffw_layer_1.output_min": "model-00001-of-00002.safetensors",
         "audio_tower.layers.2.feed_forward1.ffw_layer_2.input_max": "model-00002-of-00002.safetensors",
         "audio_tower.layers.2.feed_forward1.ffw_layer_2.input_min": "model-00002-of-00002.safetensors",
         "audio_tower.layers.2.feed_forward1.ffw_layer_2.linear.weight": "model-00002-of-00002.safetensors",
         "audio_tower.layers.2.feed_forward1.ffw_layer_2.output_max": "model-00002-of-00002.safetensors",
         "audio_tower.layers.2.feed_forward1.ffw_layer_2.output_min": "model-00002-of-00002.safetensors",
         "audio_tower.layers.2.feed_forward1.post_layer_norm.weight": "model-00002-of-00002.safetensors",
-        "audio_tower.layers.2.feed_forward1.pre_layer_norm.weight": "model-00002-of-00002.safetensors",
+        "audio_tower.layers.2.feed_forward1.pre_layer_norm.weight": "model-00001-of-00002.safetensors",
         "audio_tower.layers.2.feed_forward2.ffw_layer_1.input_max": "model-00002-of-00002.safetensors",
         "audio_tower.layers.2.feed_forward2.ffw_layer_1.input_min": "model-00002-of-00002.safetensors",
         "audio_tower.layers.2.feed_forward2.ffw_layer_1.linear.weight": "model-00002-of-00002.safetensors",
@@ -1032,10 +1032,6 @@
         "language_model.model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.15.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.15.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.15.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.15.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.15.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1043,9 +1039,6 @@
         "language_model.model.layers.15.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.15.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.15.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.15.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1067,10 +1060,6 @@
         "language_model.model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.16.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.16.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1078,9 +1067,6 @@
         "language_model.model.layers.16.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.16.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.16.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1102,10 +1088,6 @@
         "language_model.model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.17.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.17.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1113,9 +1095,6 @@
         "language_model.model.layers.17.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.17.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.17.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1137,10 +1116,6 @@
         "language_model.model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.18.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.18.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1148,9 +1123,6 @@
         "language_model.model.layers.18.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.18.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.18.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1172,10 +1144,6 @@
         "language_model.model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.19.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.19.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1183,9 +1151,6 @@
         "language_model.model.layers.19.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.19.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.19.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.2.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.2.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1242,10 +1207,6 @@
         "language_model.model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.20.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.20.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.20.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.20.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.20.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1253,9 +1214,6 @@
         "language_model.model.layers.20.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.20.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.20.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.20.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1277,10 +1235,6 @@
         "language_model.model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.21.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.21.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1288,9 +1242,6 @@
         "language_model.model.layers.21.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.21.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.21.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1312,10 +1263,6 @@
         "language_model.model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.22.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.22.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1323,9 +1270,6 @@
         "language_model.model.layers.22.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.22.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.22.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1347,10 +1291,6 @@
         "language_model.model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.23.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.23.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1358,9 +1298,6 @@
         "language_model.model.layers.23.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.23.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.23.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1382,10 +1319,6 @@
         "language_model.model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.24.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.24.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1393,9 +1326,6 @@
         "language_model.model.layers.24.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.24.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.24.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1417,10 +1347,6 @@
         "language_model.model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.25.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.25.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1428,9 +1354,6 @@
         "language_model.model.layers.25.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.25.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.25.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1452,10 +1375,6 @@
         "language_model.model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.26.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.26.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1463,9 +1382,6 @@
         "language_model.model.layers.26.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.26.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.26.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1487,10 +1403,6 @@
         "language_model.model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.27.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.27.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1498,9 +1410,6 @@
         "language_model.model.layers.27.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.27.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.27.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1522,10 +1431,6 @@
         "language_model.model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.28.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.28.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1533,9 +1438,6 @@
         "language_model.model.layers.28.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.28.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.28.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1557,10 +1459,6 @@
         "language_model.model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.29.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.29.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1568,9 +1466,6 @@
         "language_model.model.layers.29.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.29.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.29.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.29.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.3.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.3.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1627,10 +1522,6 @@
         "language_model.model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.30.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.30.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.30.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.30.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.30.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1638,9 +1529,6 @@
         "language_model.model.layers.30.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.30.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.30.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.30.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1662,10 +1550,6 @@
         "language_model.model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.31.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.31.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1673,9 +1557,6 @@
         "language_model.model.layers.31.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.31.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.31.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1697,10 +1578,6 @@
         "language_model.model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.32.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.32.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1708,9 +1585,6 @@
         "language_model.model.layers.32.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.32.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.32.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1732,10 +1606,6 @@
         "language_model.model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.33.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.33.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1743,9 +1613,6 @@
         "language_model.model.layers.33.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.33.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.33.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1767,10 +1634,6 @@
         "language_model.model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.post_per_layer_input_norm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.34.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.34.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
@@ -1778,9 +1641,6 @@
         "language_model.model.layers.34.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.34.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.34.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
-        "language_model.model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
         "language_model.model.layers.4.layer_scalar": "model-00001-of-00002.safetensors",
         "language_model.model.layers.4.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
@@ -1992,6 +1852,8 @@
         "language_model.model.layers.9.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
         "language_model.model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
         "language_model.model.norm.weight": "model-00001-of-00002.safetensors",
+        "language_model.model.per_layer_model_projection.biases": "model-00001-of-00002.safetensors",
+        "language_model.model.per_layer_model_projection.scales": "model-00001-of-00002.safetensors",
         "language_model.model.per_layer_model_projection.weight": "model-00001-of-00002.safetensors",
         "language_model.model.per_layer_projection_norm.weight": "model-00001-of-00002.safetensors",
         "vision_tower.encoder.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",

processor_config.json

@@ -29,6 +29,27 @@
   },
   "image_seq_length": 280,
   "processor_class": "Gemma4Processor",
+  "video_processor": {
+    "default_fps": 2.0,
+    "do_normalize": true,
+    "do_rescale": true,
+    "image_mean": [
+      0.0,
+      0.0,
+      0.0
+    ],
+    "image_std": [
+      1.0,
+      1.0,
+      1.0
+    ],
+    "max_soft_tokens": 70,
+    "num_frames": 32,
+    "patch_size": 16,
+    "pooling_kernel_size": 3,
+    "rescale_factor": 0.00392156862745098,
+    "video_processor_type": "Gemma4VideoProcessor"
+  },
   "feature_extractor": {
     "feature_extractor_type": "Gemma4AudioFeatureExtractor",
     "sampling_rate": 16000,

tokenizer_config.json

@@ -17,50 +17,72 @@
     "<|video|>"
   ],
   "image_token": "<|image|>",
+  "is_local": true,
+  "local_files_only": true,
   "mask_token": "<mask>",
   "model_max_length": 1000000000000000019884624838656,
+  "model_specific_special_tokens": {
+    "audio_token": "<|audio|>",
+    "boa_token": "<|audio>",
+    "boi_token": "<|image>",
+    "eoa_token": "<audio|>",
+    "eoc_token": "<channel|>",
+    "eoi_token": "<image|>",
+    "eot_token": "<turn|>",
+    "escape_token": "<|\"|>",
+    "etc_token": "<tool_call|>",
+    "etd_token": "<tool|>",
+    "etr_token": "<tool_response|>",
+    "image_token": "<|image|>",
+    "soc_token": "<|channel>",
+    "sot_token": "<|turn>",
+    "stc_token": "<|tool_call>",
+    "std_token": "<|tool>",
+    "str_token": "<|tool_response>",
+    "think_token": "<|think|>"
+  },
   "pad_token": "<pad>",
   "padding_side": "left",
   "processor_class": "Gemma4Processor",
   "response_schema": {
-    "type": "object",
     "properties": {
+      "content": {
+        "type": "string"
+      },
       "role": {
         "const": "assistant"
       },
       "thinking": {
         "type": "string"
       },
-      "content": {
-        "type": "string"
-      },
       "tool_calls": {
-        "x-regex-iterator": "<\\|tool_call>(.*?)<tool_call\\|>",
-        "type": "array",
         "items": {
-          "type": "object",
           "properties": {
-            "type": {
-              "const": "function"
-            },
             "function": {
-              "type": "object",
-              "x-regex": "call\\:(?P<name>\\w+)(?P<arguments>\\{.*\\})",
               "properties": {
-                "name": {
-                  "type": "string"
-                },
                 "arguments": {
+                  "additionalProperties": {},
                   "type": "object",
-                  "x-parser": "gemma4-tool-call",
-                  "additionalProperties": {}
+                  "x-parser": "gemma4-tool-call"
+                },
+                "name": {
+                  "type": "string"
                 }
-              }
+              },
+              "type": "object",
+              "x-regex": "call\\:(?P<name>\\w+)(?P<arguments>\\{.*\\})"
+            },
+            "type": {
+              "const": "function"
             }
-          }
-        }
+          },
+          "type": "object"
+        },
+        "type": "array",
+        "x-regex-iterator": "<\\|tool_call>(.*?)<tool_call\\|>"
       }
     },
+    "type": "object",
     "x-regex": "(\\<\\|channel\\>thought\\n(?P<thinking>.*?)\\<channel\\|\\>)?(?P<tool_calls>\\<\\|tool_call\\>.*\\<tool_call\\|\\>)?(?P<content>(?:(?!\\<turn\\|\\>)(?!\\<\\|tool_response\\>).)+)?(?:\\<turn\\|\\>|\\<\\|tool_response\\>)?"
   },
   "soc_token": "<|channel>",