diff --git a/console.log b/console.log new file mode 100644 index 0000000000000000000000000000000000000000..6df31f65219871d476536dec544f4dd9423b6700 --- /dev/null +++ b/console.log @@ -0,0 +1,53 @@ +None of PyTorch, TensorFlow >= 2.0, or Flax have been found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. +None of PyTorch, TensorFlow >= 2.0, or Flax have been found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. +VLM initialization starting ... +VLM initialization completed. +>>> Query: Why is the sky blue? +Assistant: The sky appears blue because of a phenomenon called Rayleigh scattering, named after the British physicist Lord Rayleigh, who first explained it in the late 19th century. + +Here's what happens: + +1. **Sunlight enters Earth's atmosphere**: When sunlight enters our atmosphere, it encounters tiny molecules of gases like nitrogen (N2) and oxygen (O2). +2. **Scattering occurs**: These gas molecules scatter the light in all directions, but they scatter shorter (blue) wavelengths more than longer (red) wavelengths. This is known as Rayleigh scattering. +3. **Blue light is scattered**: As a result, the blue light is scattered in all directions, making the sky appear blue to our eyes. +4. **Red light continues to travel**: The longer wavelengths of light, like red and orange, continue to travel in a straight line, reaching our eyes from a more direct path. + +This is why the sky typically appears blue during the daytime, especially during clear weather. At sunrise and sunset, the light has to travel through more of the atmosphere, which scatters the shorter wavelengths even more, making the sky appear more red or orange. + +It's worth noting that the color of the sky can vary depending on atmospheric conditions, like pollution, dust, and water vapor. For example, during a severe thunderstorm or in areas with high levels of pollution, the sky can appear more gray or hazy. + +So, to summarize, the sky appears blue due to the scattering of sunlight by tiny molecules in the atmosphere, which scatters shorter wavelengths (like blue light) more than longer wavelengths (like red light). + +Number of generated tokens: 327 +TTFT: 0.21s +TPS: avg=19.64, quantiles=['20.03', '19.63', '19.58', '19.35'] +>>> Query: Summarize +Assistant: The sky appears blue due to Rayleigh scattering, where tiny molecules in the atmosphere scatter sunlight, scattering shorter (blue) wavelengths more than longer (red) wavelengths. This scattering occurs when sunlight enters the Earth's atmosphere and encounters gas molecules like nitrogen and oxygen. As a result, blue light is scattered in all directions, making the sky appear blue to our eyes. + +Number of generated tokens: 74 +TTFT: 0.13s +TPS: avg=19.10, quantiles=['19.12', '19.10', '19.08', '19.07'] +>>> Query: Tell me a story based on this +Assistant: Once upon a time, in a small village nestled between two great mountains, there lived a young girl named Luna. Luna was fascinated by the sky and spent most of her days gazing up at it, trying to understand its secrets. + +One day, while exploring the village, Luna stumbled upon an old wise man named Atlas. Atlas was a master of the stars and the secrets of the universe. He had spent his life studying the movements of the planets and the behavior of light. + +Luna approached Atlas with a curious mind and a burning question: "Why is the sky blue?" Atlas smiled and began to tell her a story. + +"Long ago," Atlas said, "the sky was not blue. It was a deep, fiery red, like the embers of a dying fire. But one day, a great storm swept through the land, bringing with it tiny particles of dust and gas. These particles danced in the air, scattering the light of the sun in all directions." + +Luna's eyes widened with wonder as Atlas continued. "The blue light, being the shortest and most energetic of all, was scattered the most. It bounced off the particles and filled the air with a brilliant blue hue. And so, the sky became blue, a reflection of the beauty and wonder of the universe." + +Luna listened, entranced, as Atlas told her of the ancient Greeks who had first discovered the secret of the blue sky. She learned of the great scientists who had studied the behavior of light and the tiny molecules that scattered it. + +As the sun began to set, casting a warm orange glow over the village, Luna looked up at the sky and saw the blue hue for the first time. She felt a sense of wonder and awe, knowing that the sky was not just a simple reflection of the sun's light, but a complex and beautiful dance of particles and light. + +From that day on, Luna spent every clear night gazing up at the sky, searching for the secrets of the universe and the magic of the blue light that made it all possible. And Atlas, the wise old man, watched over her, guiding her on her journey of discovery and wonder. + +The story of the blue sky became a legend, passed down through generations of villagers, a reminder of the beauty and wonder of the universe and the magic that lay just beyond the edge of our everyday world. + +Number of generated tokens: 476 +TTFT: 0.13s +TPS: avg=18.45, quantiles=['18.80', '18.50', '18.30', '18.09'] +>>> WARN:starting syslog with prefix MLA-RT +~MLALogger: logger is closed diff --git a/devkit/precision.json b/devkit/precision.json index 5323756c08266a96d4cacdd733daf913c7e77c64..dfbb3f209309b40913ad5e1bfd070b1571a08e34 100644 --- a/devkit/precision.json +++ b/devkit/precision.json @@ -2,357 +2,357 @@ { "part": "group_pre", "idx": 0, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 1, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 2, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 3, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 4, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 5, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 6, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 7, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 8, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 9, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 10, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 11, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 12, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 13, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 14, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 15, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 16, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 17, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 18, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 19, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 20, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 21, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 22, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 23, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 24, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 25, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 26, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_pre", "idx": 27, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 0, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 1, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 2, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 3, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 4, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 5, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 6, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 7, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 8, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 9, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 10, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 11, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 12, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 13, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 14, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 15, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 16, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 17, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 18, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 19, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 20, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 21, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 22, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 23, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 24, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 25, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_post", "idx": 26, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 0, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 128, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 256, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 384, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 512, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 640, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 768, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 896, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 1024, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 1152, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 1280, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 1408, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 1536, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 1664, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 1792, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "group_cache", "idx": 1920, - "precision": "A_BF16_W_INT4" + "precision": "A_BF16_W_INT8" }, { "part": "single_pre", diff --git a/devkit/vlm_config.json b/devkit/vlm_config.json index dc0bfaf0fd2bc6d55a02c4e5a08f40239cf4af7e..588ec4f5e0596883a5e6437f1a3d9ce185039899 100644 --- a/devkit/vlm_config.json +++ b/devkit/vlm_config.json @@ -58,6 +58,7 @@ }, "pipeline_cfg": { "system_prompt": null, + "chat_template": null, "max_num_tokens": 2048, "input_token_group_size": 128, "input_token_group_offsets": [ diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token0_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token0_stage1_mla.elf index 0e763e8ecd5b53862d08d5b48ee085e34955eb36..92cdc00b56b21af4fe509f9a502a02b5a7014563 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token0_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token0_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:1cc7cf91d8b00fa25199b4f59a06cf10a727efcd01927b4153cde5a0af265198 -size 3416472 +oid sha256:ca700b22e336c83e71d880148fc48790736e2aa79d25b1a25c9b91ace5b46297 +size 3401432 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1024_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1024_stage1_mla.elf index e0da785115850125a1e23fb272e9e4d53c6992fe..4461a4372773bb3d39d05ba4d323604fe0fc6861 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1024_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1024_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:db832a41ec987b93a05e744f4a4a2e8010ba9a4eefe0bb831025689612bdc333 -size 12117008 +oid sha256:5a46169a6ef8fccd0d039a5b0f9bdfb93f26a71d187492796be4b1ab90aaeab9 +size 10845272 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1152_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1152_stage1_mla.elf index 3b110eedc6adb1a48bde29750baf12af861e78ea..4c5a4090da95c583298285558a9b6f559f57400a 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1152_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1152_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:aab64711709a94a413709820fadc76cdee8520ff183a717dc765a4bf298d5eda -size 24802248 +oid sha256:a5ef188745c39367a33ab9063e2b09fca4e11dea0c381b8818cddf868d35b119 +size 13135256 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1280_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1280_stage1_mla.elf index 2ce09f2ec785b5687e525ab5a437c0621bf849af..cc495997f5f4e20681058ef17adf19c628a99eb4 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1280_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1280_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:869beb962c1b53981f6eff6d2bb73df3b482c6079215c1edfd227eb59256f896 -size 18557320 +oid sha256:6f7de46dcae36af5daeaec2d934c045ee2136188870bfc0156b1cb367a275031 +size 12327728 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token128_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token128_stage1_mla.elf index 2b2bbf94e16571f5a2f3f032aa705dba656e89c3..44d18b74dbd49633b95d0263009914e103018f19 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token128_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token128_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:f7038f9fd20fd49180a509da77a73b4ff163a3043b966d53f6c06d440f3487d5 -size 3548848 +oid sha256:70ebb28ee3f4c51f24ecdceca2c3a863f0f62ca4a1565c420d55525614c6b2c5 +size 4233104 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1408_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1408_stage1_mla.elf index e8f4056190164a5b88a170b483ef3e312c8e3038..8c0b146069c94258ddb45cebab2dc7f5eff0e087 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1408_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1408_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:dcd4f4b853edd384d58b7e9ad7c9a0116f7e9c6c9ee5f86fda64cb4fe7777409 -size 13150896 +oid sha256:896cb3c3aae95b8e7ab7c1f74b902b7c9eb62eaf21a8a3fa48a958b575a3b9cc +size 15098336 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1536_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1536_stage1_mla.elf index 3dccd1659639f0124585a31886257d0c5f3d22d5..9317e343622dc3633cb622fd8dc596a1f700c98a 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1536_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1536_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:282553c8e06da07e99b592e6bdda256cf803800262c426b57991a9f8a3ebc4d6 -size 12214048 +oid sha256:405fa5b6f5ce843d3f6419891b1140c44be8d533043a09536cb144aeb9ecae5a +size 12033840 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1664_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1664_stage1_mla.elf index 7f0ffbee868b903bb3c1348ff60c36c395f5eadb..bfd9caac8dc0bdeb1ad85e0bad73664eeced0032 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1664_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1664_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:e7c2b11ea80d07aab5df9b6e14b44f8c482c756abdb5780de6ce1287345c17c6 -size 24761864 +oid sha256:2de1cab87bdc81a223154d3ecbb303690087edaa41fb27049df94cb9265b4a1c +size 12084936 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1792_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1792_stage1_mla.elf index 86a40c25fa7ad900185b8141617883b56f5bb2c1..f5ca45071cb0263115843470e73bec491856da64 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1792_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1792_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:b7a0ff25730f3ed0b8e784a265b0cea166b08060233fdb93463b31b4508073a0 -size 22043496 +oid sha256:0e1848620773150da324539516d61a425c74f08fe2ebdae20f0176d9f93c609c +size 13031456 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1920_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1920_stage1_mla.elf index 0a65acae9b0f71c21666d4d1760a857ceb7ed6dd..105c74b7ffa188377d2fff2e401c5c52cabe5dbc 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1920_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token1920_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:e8c0b6301b61bab3594b1240772454561b361d5115130c242e65d18fbdcd5b0b -size 15465568 +oid sha256:29f68808d78245db93cb412cf22ceabdf9df91707dac3280af7e237b6eee85ab +size 13259552 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token256_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token256_stage1_mla.elf index 8857d5af3c40436fe3f15964f5f16f96fb153be3..42da47bd002b9901d4878f356ec7d7c5a40616cc 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token256_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token256_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:37e3ee9f6718ce604807796612ffc41a1234b2c1a787e5c0cd043a05da86cbb2 -size 3933856 +oid sha256:7ba56b4e2ef804240fa93cc37da5d0193bd451c22f2b89cf3031a47a3ea99e9c +size 4396720 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token384_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token384_stage1_mla.elf index 7abd1b5dc1a641e67e6ff46b8313d9256a182fd3..94be1c993d1eee445c0fd139c5d53175004f9591 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token384_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token384_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:673c505346c162d9e9adaa05825ccedbebe84d2c0efb13cf7d86bb3a75cd1d36 -size 4238864 +oid sha256:20bc01829749925f9eb22110a3fbf35236524a171132b7841303690d4de224e0 +size 4621064 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token512_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token512_stage1_mla.elf index a238b650e7cbb1b0e94d03b874b35d05784b8c10..bcfe712df44d95b989b25675bded2f75c2d3517f 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token512_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token512_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:95246069dcd64bc618c8597983f842f167986a70c6ec78c6ab61f4ec7a1c16d4 -size 6740760 +oid sha256:ac8590821dbdb8ce3a9faff797fa4180bb57cb64e16545ce657437298411111f +size 5901152 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token640_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token640_stage1_mla.elf index fe17d8570b421bd5a2131c60b53df8b6f671d269..42e30f23767d39e33a762b8eadce4d4d50ee44f8 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token640_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token640_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:826cd2e1623f7635955c24b973d980b357c71005d0bf7db3a99d47cee4d5a808 -size 6221152 +oid sha256:5d3b95e893a63f13b38645496a47df5a14e1b78d909634ee121dabd08f889ff5 +size 5997512 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token768_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token768_stage1_mla.elf index ab7c9a4abd47950dbf15ce5f90c55e26074ca8c0..3b872fe5dae962a9e2525f204d852a28f56769d8 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token768_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token768_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:bafd240c169ae9fa2c555d319132cca4ef030c2fe1bf60b525bcc29de17fcf5c -size 9033224 +oid sha256:f97a8933bafba66a4fe026ff39ac46427212672d1f9090c01918852d7bdd32e9 +size 6072704 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token896_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token896_stage1_mla.elf index 787b820ce0a5f80eeb05b1b20f013ed8725dbaea..b2ce83f94824ee772d0d1284e910c3038813d38b 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token896_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_cache_token896_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:985e4651fab8632aa4cfbefc3013b59e68f1378330eb64d6d07021565e8f96cd -size 8640680 +oid sha256:dc8d79b37a359a6d6df883eb6f22e4eaf7566a9dbb73f9692a89db15accab2fb +size 7383576 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer0_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer0_stage1_mla.elf index ceb9e3f0c25affb29673bf4d037ef5c1cab2ed16..e3a9a18b7d2726f2340633b4912d9fe6d3c21c99 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer0_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer0_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:386b495f68c8b1b07b934bbc4e284c72a187e6abc62a020bd86ca8177e29beb7 -size 112493256 +oid sha256:89bb0d4f3b764e057e61c189a21c2d011d07160c66ef4513cb99ed391d882538 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer10_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer10_stage1_mla.elf index f8b60e4284aff213b030c4c42dff42db85deb206..322ac6f700ede96c096105845ced97c834b43125 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer10_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer10_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:9bdf7f36c3eb457683fad82cda39476a47492927d0713f4c2613b8206d09233a -size 112493256 +oid sha256:4fe12defe5da30929f5b14abf21ada4dfe4801c6393a306274f657b4ca5ef9d0 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer11_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer11_stage1_mla.elf index d7fc40d134f6e5c9e473a3b98e12a7ea1666bc55..350073e503fe3543edad043160b2ec3ff53029c8 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer11_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer11_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:7080d6c88bcb8e8265df94b210e3a293d9a20440daf4d8b11e9de3cd2c1be01d -size 112493256 +oid sha256:c877069bef2789388b91bf6f93115493e3c0c8a34903e98572f5f9d9ff7cc689 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer12_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer12_stage1_mla.elf index 26d72cc580e5e212acdff477e18949235dfe5a69..6c1c94668d32e4851986d2a238d3e88e4009959f 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer12_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer12_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:a524af24380b37d83dd933139d0ab61902c9ef1602e0d2e3d4de51030084a5b7 -size 112493256 +oid sha256:d98da92ff4256586df904d73d066564d85e24d45ba5659a3f4bb1e693a09139d +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer13_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer13_stage1_mla.elf index 17c44f966a8774f6d68ebb8baf3b8552a8eef159..3dcac082a25c6d282918006af5b4cab18ab1184e 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer13_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer13_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:48ff629c95f99e5f224bd670dc5f63f6c03d2c53261a2a670d687ebd8b20b472 -size 112493256 +oid sha256:a855e8efaa8f1d7336306a604f1a68217b16409d5e652b18c83e1357e3ef4568 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer14_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer14_stage1_mla.elf index 0c93d26d69ea3644825de54ee351a80d3b284d00..59c06ed468f3bbedf96adb05de54e99532af2209 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer14_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer14_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:c6844439ff9df34e86df6bfe78da5831523e3e7c2bfd192b68a385ca4afee68d -size 112493256 +oid sha256:dbfacc6c8746d483c117a5283cbb5d5ce64ebadaf75bc17f36663302e207ee56 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer15_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer15_stage1_mla.elf index dbc8c1d9fb1af320538d4e0a1cd7c72a4a3c3c88..50ca170c7c23385a9479f4c300a0e3a4f75d9fa2 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer15_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer15_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:23c9d6ddb0f2a72df1f491eb3007b331dff2a9bc2b40e307dec818ddebfe75b5 -size 112493256 +oid sha256:89e5a35fe6b214ee4e323b77142467f890cafcd7f19e4e9fe629bf439a2772db +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer16_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer16_stage1_mla.elf index bc5da210204566b6a1dafc50bd67dfeb1b93e9ec..212110d102b4752d40304bc2180f1ff27d4b3911 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer16_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer16_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:647dcee9e256859a92a540e0f91327dcb38a5a2b52ca3012cf2666c73495c637 -size 112493256 +oid sha256:90b6727a850ec028597ad824043122d0071471dc9ec18cb653b53a47f61a5195 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer17_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer17_stage1_mla.elf index 67b4f7e4fcdd23a45a765e69a35165b105a46c6d..f7c0409561f15bd786b8d86e434689f4fc84399e 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer17_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer17_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:131ea259b9973d4b567272461b6b40c07a5dfe24dd329fdd56d3ecc43699b9e2 -size 112493256 +oid sha256:2b46df5861eea55437dae07b00452cc9a0c3c96dcceefbfa33e85b47946d0b3e +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer18_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer18_stage1_mla.elf index 13509fb414d309269fd563836e91b7672e77d9f4..faf4f90ffa4339cf25e1b633095eee86c24f9320 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer18_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer18_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:a0b56160b5718c82c9056f0c651db2485a2661fba2f817e8c9a87d8be574cb37 -size 112493256 +oid sha256:06372bb3504f5ef095645b2f5c2c477f4b4c79d4447e22303541b7cae8253b3b +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer19_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer19_stage1_mla.elf index fdae756473fbe199b99f7fc2c4aba4dc0f8965e9..07b1b93378254f70bcd0ae5160bc5155dc298f0c 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer19_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer19_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:03b2c8e5d42c721d432ec4452eba1effe28bec3ed1abb600f73b7f5a623a39d2 -size 112493256 +oid sha256:75a855b5426a45f291592f11a3af059f54cc5417b8b8ee9b7a209dc8cb3f3893 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer1_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer1_stage1_mla.elf index 5017b8fd5681fae70bcbbe178851d837a545fbe0..6d6b7781864885a3eac88cb4024e434e01f8f04c 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer1_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer1_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:5cb13102a1cb85ed42c3d09864167ba1ded551a453d74d376a6095d55f7d2d2f -size 112493256 +oid sha256:8f17bb9d7dc55a8acdd9ad0a577bbd5091b1f97cd80d61c404148b8edb40a3d3 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer20_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer20_stage1_mla.elf index c025b3303e57ee2d8a4205d0142fd6da9f422d68..35d8e19b12b9f1d91b3820d9ccc0ab2c6a0622c7 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer20_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer20_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:1d260601b3507a9a064d3cbb5cd6fa0700db1eee2959ca40b83c28b7fa685d48 -size 112493256 +oid sha256:986b0496a5ac6e0f7137d1b92b3cd88b817167f2891b771a26b0cf69cd2e0f9e +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer21_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer21_stage1_mla.elf index a29f683b521e57174c158e49f22abcf5f0b734b6..28a60d805e6e3c8134d661b9cc87dc05eb7853ff 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer21_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer21_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:658ccb8b5e575e98608f491c7d1dfdd0e3c821deca0f39b3d495913124f61781 -size 112493256 +oid sha256:503cbf9435c4c674855ecba6a158fe70dbb82f93bf908abb80e25274aa77c867 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer22_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer22_stage1_mla.elf index 57dd46bb84adf4357dc69d3baed14a67af9bec7d..9eec6b166677e01bd7ce3d3332c32b7cede1e594 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer22_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer22_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:82a0fde8d6655f18ecbc5c834c05f8010f9c8c85a24078118adde6410cf5bba1 -size 112493256 +oid sha256:f54c127c779643907294d47887a26ada15aa14eece2ad7008155e94c87bc1c49 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer23_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer23_stage1_mla.elf index cd0d7a35b7af62dd785301130050a9054c7b3efc..18c9e1f75a3b6052595b90ce274dada455fa4db1 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer23_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer23_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:b5a6b44def88d0fa02f9c32b3bdfb38e07cc7ddf9032d8967178c59ef79532ee -size 112493256 +oid sha256:db38a4c71ceede23f9ef22c02ef34f1afefe09398f03a8b0c1a1f2bcd53367c0 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer24_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer24_stage1_mla.elf index 8308af8f10ae4f08675d3162dbe1038aef453b19..cd4089f1312a1fdfd74cc088d611cffb6890e0e3 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer24_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer24_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:dd59a2437534585cf6c60ad14a0a37249ea0b2e3c691a3917bbc514466dc52e2 -size 112493256 +oid sha256:94ce53f2e040c14c0e87aa92d227057e5a9825478ea5ed99bdb597aea5a9c944 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer25_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer25_stage1_mla.elf index 860df33f4933bb30277308442173708e5f544e48..928b4e17a90e87a1c179f3cf7335938107f74e66 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer25_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer25_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:6bbce6425fd7d9d7a72338132bbebe97530a861c2988060ead26428065191abb -size 112493256 +oid sha256:5e5d0aafb478ae9a08979c1305eb0a7e0929a84714f710b59f3b0e809e77a19a +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer26_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer26_stage1_mla.elf index bb1af13b67c0b2b343d6bacadf1eba8da4868bd7..5d6b17669bd970078200dd72103f71f9d58c1218 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer26_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer26_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:d433e27a321a0156c44081a7af43285b6a33e2a688ef88967b846117511bd47a -size 112493256 +oid sha256:a4687c78ab9fd4a075712d097ed63d4bfbf9a7504f3a43e4d2eb4fde1eac2e5b +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer2_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer2_stage1_mla.elf index c23a5590d168a9300b974c05061779a849c1914e..84985f617a1ad3d332a60695c2e5fe1c33286ede 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer2_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer2_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:34e569bce07cd81ef54b1bf627b2ecc75b4c7f245198011acff9319245e94749 -size 112493256 +oid sha256:2cda2ae33cfb044af3071daf7dba0daeb0b626482e2dd8eb761d294317764ba6 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer3_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer3_stage1_mla.elf index 1cc9ce0b194badd9b599019f92e4bb3d2f065872..643974c57a77c6139a2961407af6866fb4efe0f7 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer3_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer3_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:7539d76cde202f7346b3c0ebf55d408d68c9b00729aae8ead3faf0bde7f4ff41 -size 112493256 +oid sha256:8e14ddbe272e944c4a9983752cb998e5659b71d0e032875feca4b77afe595579 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer4_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer4_stage1_mla.elf index 45ff5d75500e782b031778cffaadd2ba507122b8..3996784a7fd79809e4fc5ad70531ce12b04ce213 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer4_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer4_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:551fbef0e897672e18bb0e7462401f76aae4e69fb432de6e2737b74b43f9fb1a -size 112493256 +oid sha256:8b9d966a14b28ee545f25b52a16f3c74a472ebe6bded9f313a4e2186ee482564 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer5_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer5_stage1_mla.elf index 2123d1792b6252402bc829b4e670b01c32fe3fa8..2a71cd88e5837a0359edb0dd4371981c1da6b1be 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer5_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer5_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:af3fc9a0e17dd4476cd9efe062f8a39b96e83b1c145b7bec88809292bc6388cd -size 112493256 +oid sha256:242a872951efdcb3c4ded087d180a4ab78a1dbb05a852ff0bc3660cbc681923a +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer6_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer6_stage1_mla.elf index 5c25be963f4d2478d8d2c98b4cf5c5f1cf520f03..517560458453556f9ab3b89a77635889959f6945 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer6_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer6_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:5cbaac4cc1906da6def30c928db966632f23ad7d9e8b2c441fabc627d69d3104 -size 112493256 +oid sha256:f485ccfd524293afdfc88d8f71c8dcd299aba0733a6ea474d48d26199ff763a0 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer7_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer7_stage1_mla.elf index 2634001be4c0eda741742a769058d6f217fc7aa9..5446341506b81ed153a89365ee4c097364fd8ade 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer7_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer7_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:7e54df77cdc5060c4722d941510358f2da444976794b8b8ca9d75a6d5cbe8f54 -size 112493256 +oid sha256:9dd6b6a9054e367b7860a17fb23cccf8bd7b6bbe23b88b86bad97b8218622841 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer8_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer8_stage1_mla.elf index b1b1630bbfa6b4822c9ff25c34038618e0da46d9..c5c7e9196a54f557cb47ceb645c863e3e4428c8c 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer8_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer8_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:d4a0f55d41ecc2306ece68b917318fecb0626c9344b2e510459f8354f8cd78c5 -size 112493256 +oid sha256:117c81f47e8281d330636d2136930e016a37da02e33a4084a40c2b8a57e4bf05 +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer9_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer9_stage1_mla.elf index 33c5be48444fa05060ec134161349af34a3a39c9..01eca4d3874a28ac844013f4f6b511bbdf1b56f0 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer9_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_post_layer9_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:5471f1402da01962aa28c963467931ba0a801b8adbc164cb229ebfaac2e4aea0 -size 112493256 +oid sha256:59a0eabd082e0cc3459ace47288f654850ea093194e970d034ac42849395200a +size 97409136 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer0_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer0_stage1_mla.elf index 01126cf8ff8a495b2656492f3cb95591b8e78bb8..398f83f8ea3ab8ffdf9c2deb82079e3689bedead 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer0_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer0_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:97b14da070c40de6de2ad4edbaa97184bbb1bb8c2974662e6cf321ee7a51fa10 -size 21633608 +oid sha256:15ef21defe8ecd945fdfff0a60b70c319d2d975a1083f6ce12848a025b96fb80 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer10_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer10_stage1_mla.elf index 0bc1ce45faedc921ffd4033103c48a8cb82e7543..9ef32547eb0c6cbd8931413e960019deb4c9c791 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer10_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer10_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:6886e86e7fe352282708af41ae70200dd078dc386d5dd71877be2a776ca4acb2 -size 21633608 +oid sha256:a96ab9b78c2d0bbbb5eeeb3083469fd2aff5ee50477a1d47288aaf6467d241ec +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer11_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer11_stage1_mla.elf index e5da7759e1e72641a4aefc87fb73eacdd388b8b9..ff254b07c739895b34b92c418076fe017a627c31 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer11_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer11_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:216d12dbf134d126bef2c030c53eba9e4f371ed7c64b769299b8e56687c2b6b0 -size 21633608 +oid sha256:3978d4768b3d95f57f48c8a717f2565a627036ad8ca4e1da309fb6fe26233e76 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer12_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer12_stage1_mla.elf index 09976faad8ea739eba483945c79ed421387104e8..5d776210bf2a931b9738bb7f73baaf4316ba262f 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer12_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer12_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:4bf962d2e560fff37136b62d46015049221985aa25c2548a715316d395dcc86a -size 21633608 +oid sha256:88336f6ed59be62f94df6612a0dc2758c11a23c7f8358358346506a947a492f3 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer13_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer13_stage1_mla.elf index 914dfd114a2bc9288831a30703cf6bbf98f6f298..ac89a6a9443592a4ebccea1ba234a203a1bf4349 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer13_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer13_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:971cbf4be5a8a675223568ba250ab5e6c81478c07a5ae88c769819c0002a33d6 -size 21633608 +oid sha256:86ad1e8c899d664c9861bc9d418f0d1bc75b035094eb03d7a44ccdb1fe90be9a +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer14_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer14_stage1_mla.elf index fdc4e5f2c8fe049af8b43edbbf6d90edba13eefe..8769c1e4a92b180d16a14de4850c79b721708259 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer14_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer14_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:b8433a316feed5063b5878ffcb7b0379bb134b7a327aee8007cce627e033f066 -size 21633608 +oid sha256:5eaa7d5dee652220f8612732626dc79e6c34601b16672ab3654298de60166b3d +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer15_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer15_stage1_mla.elf index e0159e00686ea2a5634bebe5893cc7b3da9a30a5..af14a965a3b44cd7abe5d4aeec86863f000a845c 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer15_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer15_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:6d3adfa1e2a1d35a4cc94d0c46795f16303f2272f91c50d028b357b141969978 -size 21633608 +oid sha256:86fe5e5c0bd8871f32916b29ebfe6600390e6525c20445de2476e322e28fabc9 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer16_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer16_stage1_mla.elf index ed8ab0eb43ca0c9a09c74d6c607b4db0f0cdab11..b436b9a2d2e3e4258f125488279e2834ae3d64cc 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer16_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer16_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:529d022c466e5b9432003ef9d834109405f8020cba0723030c9d5f0253e62c11 -size 21633608 +oid sha256:d0cac9db392cb3dad6adf9abfb97f9668fe92cedca5d6d27baa277e0730570b6 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer17_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer17_stage1_mla.elf index b9278c494313ea8d6c4d1f98183bb879a2c60093..6cb260a531b51290064a8f65870e1ad87b33df6a 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer17_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer17_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:eb7d5e4876a945d744347144d4d963228d5eecc73c504369a16af63308dc94fd -size 21633608 +oid sha256:bc209e07936eaaf268dad3a78cdd86924c71c38854d76e46df7191301cfe3d97 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer18_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer18_stage1_mla.elf index 0102a7565691285485159129a0ba7db4eba251ba..ec7f336827731c92a33c5716f148d218cb978ec5 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer18_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer18_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:12e41e138c87e30ce78d802ab8a5b31fb2d9be4c801532fb387913c0e99c511f -size 21633608 +oid sha256:5dfc3aeade4c0d0dfdbd68458fe593082b6b563d320cd94c92cae3d6115a2b7d +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer19_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer19_stage1_mla.elf index 9095658d7470c68c5618ba1a10fc9d9c31445f98..ce79239f2ec33e5929c563fa73e8d699f13893a1 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer19_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer19_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:ae836bf5fbbf6a3755cf15f0db668435970922a678e0cf16eea7c2c856fb6712 -size 21633608 +oid sha256:087aee0423bcc4800e3b0a4468ecaa7efa15c5b8250a3748d9faf8a5709194f9 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer1_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer1_stage1_mla.elf index b422e10e53352dc0096de0f2ba231c6eb6b6f0d4..70a5df96294ab4c71d867ed261ada8021903b18d 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer1_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer1_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:4df24ce4f797228dbf44c56050702c65d0b6dc376ecb9785af02d7f2491121e0 -size 21633608 +oid sha256:02ff710ebc053e8b1cb45ab7e3ec83695de76869c4ec8e1dd5fe5d6a7322d730 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer20_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer20_stage1_mla.elf index 64ed15b338bfafdfb5d36699ab15941dc88393cf..e212fa6230396a07bf1238ec30a2ce3808bf5fd7 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer20_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer20_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:6052010dbe14d599c4032f622ca3298c22448a710837edfea5eb9ea4994e5479 -size 21633608 +oid sha256:bedf64e03b7892fa2aebed3e405f92db16bcc29e9b95bdfc7d04fd9f21fc5813 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer21_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer21_stage1_mla.elf index 19620a0076cd4a0ec19e0cda05c72bb0bb6ff9df..9be0d066206ef0df57f0021c0726e26e9d73795f 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer21_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer21_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:df0f4357bc0ec37c98215e5168f61257ac591a89ec6c516e6f1efe3978e6ffc8 -size 21633608 +oid sha256:9c27a161f42c62b028ad7b2a7b55c948a937ba95b610ec7c385f09089a60d104 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer22_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer22_stage1_mla.elf index 678dae5f8a6468dc0fe17f046ecdcd0a0f357fd7..957f2b2b8d4f32aa1d0f5209209c5670e672a7b2 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer22_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer22_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:b8a6108393d2e459c5a474e34187ef75b2f61e24e5df07aede8156f80e7f687b -size 21633608 +oid sha256:5b62692193df94a997f7a9784ec9ab0cbc8fdef657e8b8aa51fa1d2c6733ef19 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer23_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer23_stage1_mla.elf index 3bba1ddc127f21ea5b6fc89ef5e619be27014d8e..1259133eece523f367da1fc6eb921acac6d4cb2f 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer23_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer23_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:384a84624c45bf911a6dda8bfee546c802810d14edd492f43eef17a034e42c4e -size 21633608 +oid sha256:f7ff49a6b23f9968be8b5357ca18eb06d027a67b2889c88fd68d3ce6835828cd +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer24_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer24_stage1_mla.elf index 67deeccc7f91d598e76b04bb6c13071880e461f9..99126038da76975ef89013de60e3adb0155162eb 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer24_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer24_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:db0b5e843abc75d65c7775adacee36e015bdbfe9d279c7f97655eb0fbd5c6dba -size 21633608 +oid sha256:9c90a1a220a14eb343d4aec21a307f152fce93564c084ea9e6746d0773f1505e +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer25_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer25_stage1_mla.elf index 90af4f5a041e195134ba92d79fa04f7921dd978a..4329e9ae3d80252ecac065352cccf62836a9eaa6 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer25_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer25_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:cdc814cd7d431a05a8e0a36890bf72108a6aceac299fa59e162c1172e16f1dc9 -size 21633608 +oid sha256:fe67fe65afaca317d1b20759295478447d432dd6d8c6b823162fc6f69563aa38 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer26_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer26_stage1_mla.elf index 38490398276d11dcc6b9e49092aa43974769ffbe..732835223d89f42f144c149d1276a8f45a8b49ea 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer26_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer26_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:412fe6013f37e8259fe40e528c40681ee0912e3b4b93e1dce376739153207352 -size 21633608 +oid sha256:f3fec95cc7e1d3f5342e2656b828ccf80012ca4c3f1426fd0d968e42ea72225f +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer27_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer27_stage1_mla.elf index 134821bc2db8cc4a43b209189906a36e4af8a7e6..2916f3d9dee2292ddc012c86cc81e949d406c559 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer27_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer27_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:9fdbf8703e984bddc1313299cb3c673b155beccd63d7dacbd2a98b072454334a -size 21633608 +oid sha256:2ae706e2f415962a8b62e8d287a5661eb21d690dc14373494dccef16788bf034 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer2_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer2_stage1_mla.elf index d9385f9b886c5f4adf104e18129aac2ff4fa616e..91ea566dc36c0cd7a3d866db822e9ea94aee8a48 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer2_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer2_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:3432e64bebbfa3c8c6d46698a78ce2289fc4cf032408df26ddd6adeda1a60ad9 -size 21633608 +oid sha256:426b61228f3d2de2f0eb611e4372fd5956b76c744e31af798fa2c2ecc2c7c5a1 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer3_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer3_stage1_mla.elf index 79516cbbb11bf327824a826657d5d12eae905856..8af54b1bb5a19576b74ef86e0fe045985101f853 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer3_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer3_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:4e3c2b920a9b990a0da578e95f9c9ae593950d508f8dfd7e412c5670ce267b19 -size 21633608 +oid sha256:d71414192c7f1f2d6d5ec4ef64aa01f46e227ad7b01fbc0df643024677177dd5 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer4_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer4_stage1_mla.elf index ecfa8ad01c68bbacddfd103357c24b08a812a17f..46ecb2bb3aa980387a701d4e967fd0512a936cdd 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer4_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer4_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:40edffafe2b8f464228e496198077ada757ddb34ffbe3cba35412cfa440983b1 -size 21633608 +oid sha256:953273ef533bfb6bef8a16e7e751d43e8430d4dd98f643024358bfd22778d3fc +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer5_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer5_stage1_mla.elf index 92ea96295220354fee03a7482d832f740793dba5..6c785f5ea7a64130e7053d895141e056124eb99f 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer5_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer5_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:82f026b3d425b8d5137161405a27df88f4e4351a6174a52f0350afaa04478dee -size 21633608 +oid sha256:52940f7ea1b869cc3b6e696fbea38ecea8dd231bcc647c90c6f36f2794ca10d3 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer6_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer6_stage1_mla.elf index b6de21c2fb6f87bc945d1a81d0d708558e3ba4f6..72c4e304b598c2ed395c1bc1b5fc1d5d5619ccce 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer6_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer6_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:a91dd75f28a5f7366845ff686d4a22f69f99e3e914bc27dee7b9dec8bf775924 -size 21633608 +oid sha256:a0b82f98875e57929881f4739280e99c7f67ade6c55b6cf47261ffa5dba2776d +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer7_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer7_stage1_mla.elf index 16d421e25cc0a8d0fa0e11764cf87552dc8583a4..ccba9d9d94ff57d7e31bf09e6304b7ecfb8ed99c 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer7_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer7_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:79a1f577afad5f5d3357d5069db837a4b17745711044ace92026d3c3d945f766 -size 21633608 +oid sha256:619914957dc241faef6a8ab1c85fa37cf872ff77ea48d138f77c0236b10322a3 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer8_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer8_stage1_mla.elf index 2d8a5c2918304ee8e90bab27f79da3ee04ca2801..8c03cff689e31b8adab86d61bbe78d7e0a96719d 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer8_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer8_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:9c7afceab97b3da8e8ddce7daa276fd5a140b3ea31b3f3fda231fb9e65cecfb6 -size 21633608 +oid sha256:17dcc1c1bcc2672bee48b382ec7e5ff36ee4d34f2805519f4bd50005e9591462 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer9_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer9_stage1_mla.elf index b8b6073c2f0988e39f87ca3c509ea6f855b2333b..968e0a56290d7facebd0cbf066badf1bd46c948b 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer9_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n128_pre_layer9_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:e673f1cf439ca04923fd07e8c4c5c2f171bc33cc8329afc7bae3deb21f315e2b -size 21633608 +oid sha256:c7c724eb86ab3b129b21e4804ab8a01b7affff327e72288355453430221f5442 +size 21807424 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1023_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1023_stage1_mla.elf index 13bb3bac06231d58869fe1cfaa29a54378cac1c6..554cc9f5bf6af1fee4094b441670c953fc000843 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1023_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1023_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:2d889614dd7d36668b2ac1d859109655a1d1aeb9b9420a85de94a0a6241d2e37 -size 4251112 +oid sha256:0a96c1c7032afd88d367d5ee15be125b52bd633cc2102f4ad13084b5efc9e333 +size 3234752 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1151_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1151_stage1_mla.elf index 3d39206b08bf97a8a9296b90f7a0421e68632835..c6f7ca5426921dbc13b7b3bf87b3595562fe730d 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1151_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1151_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:1e6eb6b16349795e4206554110c860a80c09155cbae6ab51c4643544f7d6ad9a -size 6509224 +oid sha256:5d49aaae389efe08adc69d563a0a851dc8781260676eb82f1c0f804888c1bc19 +size 6180392 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1279_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1279_stage1_mla.elf index d5d820ecc88fc89bd2775b2bf03b8bcfbb02ee47..e890e4884d8531fa13abf9b427a0d8076fb2d447 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1279_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1279_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:c08d0516fc1057746d6d3c05f9ef0275bcc0a6dade58215150bd2a58421127cf -size 7137152 +oid sha256:822ea753804a121eba31b24adcfde5acf65b42e50a4fcf8048d1bcce24c953a3 +size 7035216 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token127_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token127_stage1_mla.elf index 41aafaa8ec0cab55668102f2ed1353ffc63db756..bf40d5a9afe870b35e172094705948534baf14eb 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token127_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token127_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:134b897e046b0ebe5bd64bf542eb8b44115f48e1b61fcb2c4e362817be5ff7ef -size 2037952 +oid sha256:48b88d136949347eef08720c89399abe59db229b57c7ab20932a0322de5708eb +size 2115264 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1407_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1407_stage1_mla.elf index 808d15f4abf3b910bc6f36372db0c8169646289a..edf65adaee8eaf070d7c745efd87c93e54acbceb 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1407_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1407_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:80341009a8fd22add5e276b245fd31d8555e44bed75ef82dd501edfb24ab6404 -size 6370208 +oid sha256:0f39394bb3d5e5c6f8a3bc9026f81f407a9e865dfbe6bad24961ddfb26f0dad1 +size 6388464 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1535_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1535_stage1_mla.elf index 5e45910a47ce3136cc381e4ffc4f61c7d0920eda..0f39b85904c5d92a2f86b9d6b2f30fcda1b7712f 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1535_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1535_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:de959ae865fdb7824ac1971011b53bf1b1f862951555248e4487f439bc313c4d -size 6968640 +oid sha256:b360921bff7f878879cff43708c6795ef09f6848570c201ee8b5e47dd95cd714 +size 7095616 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1663_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1663_stage1_mla.elf index b4a96a4faa15932bf8f78531c2872ee17410515f..642d05527ba7543737efbfa37a50ea9886849b0b 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1663_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1663_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:cf24cedc7266244cb5a666fd80965e150a4ca674e8b5acaa7b8d06ef5180ea24 -size 4687376 +oid sha256:c5a804e84826322958da0246b5efa17bfc4708f33ee08ecfa298b20150af688a +size 4860200 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1791_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1791_stage1_mla.elf index b831a69a74ec8ccdb3b5fec666ed0f096a74c300..3cdf74da3ed060b33838acb63364b82d469f9eb2 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1791_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1791_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:12f50293fa96599080001e6cf681b488acc3ee92225038e28cfdb52c63b02785 -size 6183584 +oid sha256:acdf4f6430aaa871222d1752dfecff784f40e65d1fda93ed798026738334d910 +size 5497288 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1919_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1919_stage1_mla.elf index 07cc202f779e02400d18bfbad4e662d98c8b680d..d4bcee0af7547bdcf99d553109146f7d54215757 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1919_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token1919_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:eb68249769cd39298e7b70b1058c4a44f71e20e9cddf54a5e637cd0c6da1c414 -size 5593752 +oid sha256:49f03a075e320b7852d84abbf721bb680882a8a445f3e80288a3057006f03c1b +size 5101400 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token2047_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token2047_stage1_mla.elf index 387805dd37001f3c5c11b19420980a69406fd80e..242ea1dc41c0b8cf34d3e7acb829e0c448a165c9 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token2047_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token2047_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:63d1088080146a24be55099217e9e0635e30ffe275a993b931e25ec9ad116906 -size 7249056 +oid sha256:90c7be0884683388a54a6a2e82d12c2961448b1cc65a4b20f56142cce278ea01 +size 5586152 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token255_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token255_stage1_mla.elf index 144235971aa4f315379bfb3a1a3dbfa29070a0db..ef2ad1049663126f5b21b2d00aaf2362270e78ef 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token255_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token255_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:8658d0ec23b3c7276d30d9d26a4ee73cbb5b1d27e0d999ef37a4703fe5e611c1 -size 2741624 +oid sha256:8464944ea00ec50eb1c45a4f3d0082c284efd314b2c5dd6b2c9f0bd9a77a7d70 +size 2801824 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token383_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token383_stage1_mla.elf index 4048157b301cd6220477189717adc8debfed18ad..94a2d0575fb4e77e755227d773e5410c1fed22ab 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token383_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token383_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:50be5a9863c9a642536083a3c66e8894a7428af75254203b0e0a13fbe966ef28 -size 2183016 +oid sha256:d8679d20506aeb1fd1f9f27ee5df27f49f10abaa368f4b31ff47398aea50f5e1 +size 2647744 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token511_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token511_stage1_mla.elf index 26c8d2722bfbaa4179370f8ba94670a9d6ac1d55..7aa285c430fb1fb4243b63362d779ca4ebd7d42a 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token511_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token511_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:5d1a8bb17927298a126997d444b9d0d509f2b856fe47b5f4f204b65333486fcb -size 2899256 +oid sha256:3efc505c78e50038a922d3e013c5df5a6adb9e53ded27b6d25a5d03255e46a01 +size 2953280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token639_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token639_stage1_mla.elf index a62948925fae5a95826e8a48ae97439ddc2a6279..9b3b91ed342fe0161590e9d0d1ba850981723289 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token639_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token639_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:0d95e225b6267401daaca7815c7d5b744b760cab70d0095411a4a86c24f0aaa1 -size 3592328 +oid sha256:458471a6c37194cf73cc08ef21911bf461ddd819f03688004591eae254647df7 +size 2902640 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token767_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token767_stage1_mla.elf index d49702163134401960dbbb170ab0670bab068737..5c0ae68dbd1790b608b16510070a52821c56b5fc 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token767_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token767_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:85cd19eed36c410f6d4dc255d3ff8db65c3b5a3d50ed986c8350efc713a53364 -size 3990144 +oid sha256:469d9e7a5bc2bc783c2d98f599198e2e14a22ce84499dd136d15c9e7b733504d +size 3121856 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token895_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token895_stage1_mla.elf index adab5a645558f9a37dc514010154b715cddff9a8..c93c9ad047dae6d2948b8754cc8b4fd44d612bbb 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token895_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_cache_token895_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:c18e81f7bb806a445c44da0fc6bfa7ea4b01ea6dd23fa79d47f684457059a6de -size 4012752 +oid sha256:0b39748700d3142b5099c6897341ac3d5385da60bc04776857a01de8262bc6ff +size 3005544 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer0_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer0_stage1_mla.elf index dd186b889f78fbeb0800dd515098d24a7f0b2d25..d5007d66122b8d1ad6d74e04313279a6424775e7 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer0_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer0_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:12894dc16416618293057e465ad74ca49101bc7b835cc4311350b8c34f43e4de -size 10064728 +oid sha256:be063a6dbe6e3d32179e262bbba67ac22d7102ce1a42077b749fe89357c1de6f +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer10_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer10_stage1_mla.elf index 3064cc5d863bcba4d1a0f671e875262eb8c65fc2..8686d9d8019d6c48bf8a6e0d5e4a3dd4431b2d1e 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer10_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer10_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:424ea0ab9386bfe9cc065c437234d5a2a067597fda1b56bf65924f59ed9ab716 -size 10064728 +oid sha256:1159b6a66d4d7fa807dbe077a5970c3002df5b8d488e81f00d7ef696ce6d82d3 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer11_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer11_stage1_mla.elf index 362a1d200bbaacb9d9cea05dc7a506eb4b45433e..63ca7fe187741220ffa309371ab4f171573c97c9 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer11_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer11_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:635fdba20eac4cb47dea66c68ca223dc69cef026d8d0ca01c3daee7f8f9878fe -size 10064728 +oid sha256:b5119cb2a84a33497a565f0e78a5cd671362863b9d258d809b98ddf4ad787aa8 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer12_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer12_stage1_mla.elf index 3c7d70d3740b54a0d9abc38c94895eeec4227901..613f3d966c21c87870ed1eebd0ddf88b7fd7c867 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer12_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer12_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:53dda489830c68eb1de4d813543b319731daf3b201e78f3337181bd45f742eac -size 10064728 +oid sha256:fdd22ebf9ee5c7f793f29d807c2bc02d80311bf2cfbf85ec95e6bec34b4d38c6 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer13_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer13_stage1_mla.elf index 9ea5f370faf4d0a35e65e359c61533e8d734425d..579c24633cb01c85c90c1a0b2db1523e4458e3cf 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer13_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer13_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:616736cc6542f66c8a903bfaefa5f3c9ea1e4ff058f648d84f5ca96be3545310 -size 10064728 +oid sha256:7f3fe422f1f8ad93e92e332251257964aea36f3c898f03bc3c52cf241efd1b07 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer14_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer14_stage1_mla.elf index f2d09ac12ff4b106ae2e4d80423c40032e0c33a8..ffd418404b915e141753ce7d2226ffb24b559aff 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer14_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer14_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:4d2fe3510b65c6e79ae119745b23a9a31b22af645195bc679e1deb1d97f24d2b -size 10064728 +oid sha256:872927590195a7551f2e67fbaedfb7c5acdf9c2adaa21e7e66659e62cecaee82 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer15_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer15_stage1_mla.elf index f8c9acb601d1e47a66ef50a486f763db8ba66368..b6b0f066105473da4d3817f66dbec26bed3d7e03 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer15_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer15_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:0cc1872e492e4aa9c6399c8db7ff3a42036624ac1ec952327bb26f3ff6616d03 -size 10064728 +oid sha256:35de93afa2140d357ba231c243dfd3e2b5450dcc41231a11e9fe9ae88d7f0672 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer16_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer16_stage1_mla.elf index cde9f5a7e154a40c28498fe56359e9b66d29985b..d5b42d070ceab2f78dee7220d8d45558713bc577 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer16_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer16_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:f93bf8bb636410f98cb8d45a94e0972df36554eafd39ba92e441bf75cc6f071f -size 10064728 +oid sha256:723b4fcd09a30c03576f9b78180de7610983618729594aa4ccb0e68e991026ad +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer17_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer17_stage1_mla.elf index 18083b4bcd679678cb8c94e32afc2b28810e914c..e479e47b5add7d44c27d7ebb49e4b84b0d3839cb 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer17_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer17_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:b996c7324d2014dce420ff69099fa4af4eb236e90f4e2ddd1ff3da5bfb76cb11 -size 10064728 +oid sha256:566863f863d6a422c7081d58184ac640fd55cd12698d67dc9d255f898ed8b1bc +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer18_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer18_stage1_mla.elf index d24d69b3e7ce274e54343e0103c147cc11042d6d..4a76ee7940f0f5ca47ef0fa1336cef449bfe64ea 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer18_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer18_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:25a5f62a899a435e418bc157bd47bb1fdc9624e20d6a07f3c988f661c8bd46fa -size 10064728 +oid sha256:63bcb6566d8f49dd705ba949f9c8ed6ebecfee0c9f810b3bfffb2754de906ca3 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer19_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer19_stage1_mla.elf index dfe806a1c04c36831bb06e5c46908ccbc15d9159..c57eeea439cbe98c1ea244885057a92a126ac4d5 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer19_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer19_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:61d3e4cdd8ed578352dee7ee667d385c9f71d40b04c25e57c0c1405d9f207fdf -size 10064728 +oid sha256:c817404760e4cb1a99fda4b35c4fe3513fcf8a7dfbc984c9f699409be7d82be3 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer1_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer1_stage1_mla.elf index c22d8073b3043be62ac390eaf20dbccb8e242903..6769b7e165d83bb4edb65a2c9bbdef5f42cdf336 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer1_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer1_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:945d279e5d0cebbd10ef49f3e93fa2edc6e0b3221ec5c8a1463901ad0638f6fd -size 10064728 +oid sha256:79dc7ce01272970211ac86ae322432836e91a1cc5590e980fb4c2b2eeed92d86 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer20_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer20_stage1_mla.elf index d2e3306de5b5fbcb0dd84a26342174e26c887630..13f69b5bdd7f69c57195dc48b48d96d43758729e 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer20_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer20_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:6f49b41c084172bcba5f25af4bb43fcfff40bd3c414b9e8231a309cda6b11c25 -size 10064728 +oid sha256:acc36b17808ef23fb671e17f9839fc8ec09ab39ced4f02dec7eebaff68dd5f9c +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer21_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer21_stage1_mla.elf index 34f93bbb16383f56b257dcce1a588bed320b69fd..86760d72ddc6f31a8f085b1d9fa2c4ef83bc37d1 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer21_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer21_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:7945dff183340743c9dfca0793d7741e61dd914bc622f0c3d2aadccdd80fb56c -size 10064728 +oid sha256:171ef7dd9dcf8606b546cd47bb03ea2708a5970e97ffd86eb841048030a0cdc6 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer22_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer22_stage1_mla.elf index 187007fd60c4126c70a3b09995b52ea8b52ab82e..bbcbd3343ee4a03a4f3855e0e138869ad7195c6c 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer22_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer22_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:988d5655ac5619e81c2faffc8536be0476a164f6a3d45da9e21317b14b17b9fe -size 10064728 +oid sha256:da5cea14972cfba9220dcab1e493c2c537097c2fd38da31b4bca1c34f58176cc +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer23_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer23_stage1_mla.elf index c9ab7afacd37ff5c3a4699c5b0f6f05a1140226c..2c6e8d0ce8b03fd39f33a583838e41843c813737 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer23_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer23_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:04e9638af40c56c35f95f159c79cac718703ec74fa2e619bc60e93cd7848e2d8 -size 10064728 +oid sha256:7ae045d1c0bea22b983a2db7cf2c56f7efdcbf01e75d28b8599ef567eeeff238 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer24_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer24_stage1_mla.elf index 7bb2a633b545e5c161d86aac0c8bb7483d95c349..85b0b1ea582708473a10eb782f1aac5e7c3767af 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer24_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer24_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:c06dc76e09747a201538b071e237f9924ba6faac8668ad13332cb1121fc38bdc -size 10064728 +oid sha256:0d588c3828f1767a1dd00362c3d01d3c49012d693db93cd40ffd69f1d7f2b6c8 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer25_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer25_stage1_mla.elf index 45a223d7a78017ffbc1864b930a96a617847c183..ed5d3eb5c494a0f5fe89d1ae74fa7dcac5a169ea 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer25_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer25_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:7000b20578a6bff1c5e2a8e6c8e2850dc40681674d25108dd7e36646a6be8454 -size 10064728 +oid sha256:10d0217c008d7c06425fb689de0558ce072c0b0807736b9c2bdec243f6e5c88b +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer26_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer26_stage1_mla.elf index 62ce397cff8367827f55e718114fe0b85af0b6e8..f8eed247a2edb87a0af8fbc23a386600a1197f43 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer26_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer26_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:f53f9bc6be1b58556c169cfa47c052b20170a33592a90665daa26074bcd5c9e8 -size 10064728 +oid sha256:f1e3790a7afbed1a3b4176c1756b2119e04bbaaf08ea9e14edf1d880cfcc1673 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer27_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer27_stage1_mla.elf index 8b78bd17f3d7bc179e8f7b16b2d7e18f3447d6a8..a78677aef0aa192fbd3c0a7d8ba8ecfa39294614 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer27_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer27_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:f00dcd3ed38f51b920822a16567c7a500aad2ad8b997ef3bfb618bc9a36d162c -size 10064728 +oid sha256:8613a4de35b9a179f304e61fe6961645ce89a90e4c9e32e6d3c79316afc9dd1f +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer2_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer2_stage1_mla.elf index 97a2271fc0f90942ca08f7968ac2bcaf82443421..c5f8ab7a8466c1904c2aa8bfe4239dbf73cb708b 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer2_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer2_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:3770c1f661191f2bbec4f62496e7f914f6ad1d80fbc0970f3f24c4a53bf20f0f -size 10064728 +oid sha256:31e11cfa994af20f8cc641548846a64ea6aa90600aea26a54d74248b6151ab09 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer3_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer3_stage1_mla.elf index a93c9b6ae420f9f58fafbaaddab415c9f53ff496..5a249104d81eec9e6297d8db48ae429c32250bb9 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer3_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer3_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:83f2eca4d4cea45163593559a7a7ef91bbc271e711d07744c1a01e1e04bb57cf -size 10064728 +oid sha256:9fc660a67c6716175343752813c41682cc0f204a49dee9bf26b3f5d256d938ea +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer4_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer4_stage1_mla.elf index c7fce1526ea645236bf88014a6a749da6d8d9408..d040b68647eb03f9c9eeca067634098a390f36ff 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer4_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer4_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:af6cdcf98e6f711471def236d0f22237245fcde498dd2f2b08adb5f6873338af -size 10064728 +oid sha256:4c4f51a6a3fc05efed789e0a49f1db8a8ada8badff07127ff2b5e3a22560cf12 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer5_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer5_stage1_mla.elf index 19d3e73dac052f9241738a5efb204c107cbad387..eddecb2958bd8474bc72ed1203b2ca4dc9e3fca0 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer5_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer5_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:692152741f8b5e820c7be1cbde3f336b4f2fd5d3736a739496dfb28d05486c27 -size 10064728 +oid sha256:277eb4aa281d128ac71d8cd6025f5092bd9d8d47fdc079d3c82903272ab932c4 +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer6_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer6_stage1_mla.elf index 7a056d020be16721ee4f145ba2934d86027c2408..aa378ad773d1a5667c158ae0c24528151528fe46 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer6_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer6_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:4ce1f3fa5c44a40a6b63448dec9f12d35c755624fde8b1e9333b3960e49448f3 -size 10064728 +oid sha256:33faf48890f4fc8c67085ed4d5f29385475b4931c32118ddb4035448fb4bad8f +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer7_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer7_stage1_mla.elf index 7afd0d80df943d6242bab383e8d9dadee2b0afcf..63e73c19d9fe7b5fece9a07d4a77a194837605ba 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer7_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer7_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:9e0b38e784f77dfbe575e3e78072454572c99fb38abe12a1d11ec9ea1362c8a7 -size 10064728 +oid sha256:a7df45e4fe69f85a4004a50f941af909718adeed2b4f50a54d989f2f9a994c7c +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer8_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer8_stage1_mla.elf index 3f3c9cbff81df56a8bfba167ee46b3aa744a6316..d5ba57cd5c9f9384b46288dc6deb9d8e0cd260fb 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer8_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer8_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:645df4caa411d8554171392707d2e31dbcd878697a78a1ee1c2e3a09d57f3179 -size 10064728 +oid sha256:cc3297019e6b175fdd4e97d90eb965bcd321af4be57e28a5cada8fd4f1fdd82d +size 10209280 diff --git a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer9_stage1_mla.elf b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer9_stage1_mla.elf index 30161b8daaa6ad3d5e67f97fe9dae3600f13f434..4554af4b3e0144b69ddb6213407d5fe3ff940a4e 100644 --- a/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer9_stage1_mla.elf +++ b/elf_files/models--meta-llama--Llama-3.2-3B-Instruct_language_n1_pre_layer9_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:1c3e3a0a0dde117f929c4cb2fcc7a4dd58fd6505f7494e8dd24e3c636e410565 -size 10064728 +oid sha256:6be514a696594d191cde4683e30eaced90405b5a8a699373c13db4d3e9d8856a +size 10209280 diff --git a/gen_models--meta-llama--Llama-3.2-3B-Instruct.py b/gen_models--meta-llama--Llama-3.2-3B-Instruct.py deleted file mode 100644 index 787fa1c31095b88d7032302c2c5ad6e15d3814e1..0000000000000000000000000000000000000000 --- a/gen_models--meta-llama--Llama-3.2-3B-Instruct.py +++ /dev/null @@ -1,93 +0,0 @@ -#************************************************************************** -#|| SiMa.ai CONFIDENTIAL || -#|| Unpublished Copyright (c) 2025 SiMa.ai, All Rights Reserved. || -#************************************************************************** -# NOTICE: All information contained herein is, and remains the property of -# SiMa.ai. The intellectual and technical concepts contained herein are -# proprietary to SiMa and may be covered by U.S. and Foreign Patents, -# patents in process, and are protected by trade secret or copyright law. -# -# Dissemination of this information or reproduction of this material is -# strictly forbidden unless prior written permission is obtained from -# SiMa.ai. Access to the source code contained herein is hereby forbidden -# to anyone except current SiMa.ai employees, managers or contractors who -# have executed Confidentiality and Non-disclosure agreements explicitly -# covering such access. -# -# The copyright notice above does not evidence any actual or intended -# publication or disclosure of this source code, which includes information -# that is confidential and/or proprietary, and is a trade secret, of SiMa.ai. -# -# ANY REPRODUCTION, MODIFICATION, DISTRIBUTION, PUBLIC PERFORMANCE, OR PUBLIC -# DISPLAY OF OR THROUGH USE OF THIS SOURCE CODE WITHOUT THE EXPRESS WRITTEN -# CONSENT OF SiMa.ai IS STRICTLY PROHIBITED, AND IN VIOLATION OF APPLICABLE -# LAWS AND INTERNATIONAL TREATIES. THE RECEIPT OR POSSESSION OF THIS SOURCE -# CODE AND/OR RELATED INFORMATION DOES NOT CONVEY OR IMPLY ANY RIGHTS TO -# REPRODUCE, DISCLOSE OR DISTRIBUTE ITS CONTENTS, OR TO MANUFACTURE, USE, OR -# SELL ANYTHING THAT IT MAY DESCRIBE, IN WHOLE OR IN PART. -# -#************************************************************************** - -import logging - -from pathlib import Path - -from afe.apis.error_handling_variables import enable_verbose_error_messages -from sima_utils.transformer.model import FileGenMode, FileGenPrecision, VisionLanguageModel - - -def gen_files(model_path: Path, num_processes: int, resume: bool): - enable_verbose_error_messages() - - max_num_tokens = 1024 - language_group_size = 128 - language_group_offsets = [0, 128, 256] - language_future_token_mask_size = 128 - - model = VisionLanguageModel.from_hf_cache( - hf_cache_path=model_path, - model_name=model_path.name, - onnx_path=Path(f"{model_path.name}/onnx_files"), - sima_path=Path(f"{model_path.name}/sima_files"), - max_num_tokens=max_num_tokens, - system_prompt=None, - override_language_group_size=language_group_size, - override_language_group_offsets=language_group_offsets, - override_language_future_token_mask_size=language_future_token_mask_size, - ) - - log_level = logging.INFO - for i in range(model.cfg.lm_cfg.num_hidden_layers): - if i == model.cfg.lm_cfg.num_hidden_layers - 1: - precision = FileGenPrecision.A_BF16_W_INT8 - else: - precision = FileGenPrecision.A_BF16_W_INT4 - model.gen_files( - FileGenMode.ALL, precision=precision, log_level=log_level, resume=resume, - part="single_post", part_idx=i - ) - - precision = { - "language": FileGenPrecision.A_BF16_W_INT8, - "group": FileGenPrecision.A_BF16_W_INT8, - "single": FileGenPrecision.A_BF16_W_INT4, - } - for part in ("group_pre", "group_post", "group_cache", "single_pre", "single_cache"): - model.gen_files( - FileGenMode.ALL, precision=precision, log_level=log_level, num_processes=num_processes, - resume=resume, part=part - ) - -if __name__ == "__main__": - import argparse - - # Download the HuggingFace meta-llama/Llama-3.2-3B-Instruct using the following command. - # huggingface-cli download meta-llama/Llama-3.2-3B-Instruct - - parser = argparse.ArgumentParser(description="VLM generate file arguments") - parser.add_argument("--model_path", type=Path, required=True) - parser.add_argument("--num_processes", type=int, default=1) - parser.add_argument("--resume", action="store_true", default=False) - args = parser.parse_args() - print("Arguments:", args, flush=True) - gen_files(args.model_path, args.num_processes, args.resume) diff --git a/run_models--meta-llama--Llama-3.2-3B-Instruct.log b/run_models--meta-llama--Llama-3.2-3B-Instruct.log deleted file mode 100644 index 076cb9dd419723dbf0d7bab0298091da155d339a..0000000000000000000000000000000000000000 --- a/run_models--meta-llama--Llama-3.2-3B-Instruct.log +++ /dev/null @@ -1,26 +0,0 @@ -root@modalix:/mnt/ac# python3 -m sima_utils.transformer.devkit.devkit_demo --model_path demo_models--meta-llama--Llama-3.2-3B-Instruct 2>&1 | tee console.log -I/O error(2): No such file or directory -VLM initialization starting ... -VLM initialization completed. ->>> Why is the sky blue? -Query: Why is the sky blue? -Assistant: The sky appears blue to us because of a phenomenon called Rayleigh scattering, named after the British physicist Lord Rayleigh, who first described it in the late 19th century. - -Here's what happens: - -1. **Sunlight enters Earth's atmosphere**: When sunlight enters our atmosphere, it encounters tiny molecules of gases like nitrogen (N2) and oxygen (O2). -2. **Scattering occurs**: These tiny molecules scatter the light in all directions, but they scatter shorter (blue) wavelengths more than longer (red) wavelengths. This is known as Rayleigh scattering. -3. **Blue light is scattered**: As a result, the blue light is scattered in all directions, reaching our eyes from all parts of the sky. -4. **Red light is less scattered**: The longer wavelengths of red light, on the other hand, are less scattered and continue to travel in a straight line, reaching our eyes from a more direct path. -5. **Our eyes perceive the sky as blue**: Since our eyes are more sensitive to blue light, we perceive the sky as blue, even though the red light is still present. - -Other factors can influence the apparent color of the sky, such as: - -* **Dust and water vapor**: Tiny particles in the atmosphere can scatter light, making the sky appear more hazy or gray. -* **Atmospheric conditions**: Pollution, smoke, and other atmospheric conditions can alter the apparent color of the sky. -* **Time of day and year**: The color of the sky can change depending on the time of day (e.g., sunrise and sunset) and the time of year (e.g., more intense blue during the summer months). - -So, to summarize, the sky appears blue to us because of the scattering of sunlight by tiny molecules in the atmosphere, which scatters shorter wavelengths (like blue light) more than longer wavelengths (like red light). - -TTFT: 0.11s -TPS: avg=16.88, quantiles=['17.13', '16.88', '16.84', '16.71']