diff --git a/.gitattributes b/.gitattributes index 57c2b3e1cd0f68ca4f60589223238eb1865d9d1a..d2c7a81344127bccad270d870aab2e841e9d669f 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,6 +33,50 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token0_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1024_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1152_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1280_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token128_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1408_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1536_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1664_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1792_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1920_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token256_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token2944_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token384_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token3968_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token4992_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token512_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token6016_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token640_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token7040_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token768_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token8064_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token896_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer0_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer11_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer13_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer15_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer1_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer3_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer4_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer6_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer7_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_layer9_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer10_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer12_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer14_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer2_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer5_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer8_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer10_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer12_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer14_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer2_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer5_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer8_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1023_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1151_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1279_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text @@ -44,10 +88,16 @@ elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1791_stage1_mla.elf elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1919_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token2047_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token255_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token3071_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token383_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token4095_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token5119_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token511_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token6143_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token639_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token7167_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token767_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text +elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token8191_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token895_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_layer0_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_layer11_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text @@ -72,31 +122,3 @@ elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer14_stage1_mla.elf fil elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer2_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer5_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer8_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token0_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token1280_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token1600_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token320_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token640_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token960_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer0_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer11_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer13_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer15_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer1_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer3_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer4_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer6_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer7_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_layer9_conv_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer10_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer12_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer14_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer2_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer5_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer8_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer10_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer12_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer14_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer2_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer5_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text -elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer8_stage1_mla.elf filter=lfs diff=lfs merge=lfs -text diff --git a/compile.sh b/compile.sh new file mode 100644 index 0000000000000000000000000000000000000000..4d9ead5816390d18690fb05312f4d19b2ab91407 --- /dev/null +++ b/compile.sh @@ -0,0 +1,36 @@ +#!/usr/bin/env bash + +set -euo pipefail + +# Download the source model before running this script: +# hf download LiquidAI/LFM2-350M --local-dir LFM2-350M + +MODEL_PATH="${MODEL_PATH:-LFM2-350M}" +OUTPUT_PATH="${OUTPUT_PATH:-LFM2-350M-a16w4}" + +TMPDIR="$(mktemp -d /tmp/sima.XXXXXX)" +export TMPDIR + +CONFIG_PATH="${TMPDIR}/config.py" + +cleanup() { + rm -rf -- "${TMPDIR}" +} +trap cleanup EXIT + +cat >"${CONFIG_PATH}" <<'PY' +"""Compile every LFM2-350M compiler unit with BF16 activations and INT4 weights.""" + + +def get_layer_configuration(model_properties, layer): + del model_properties, layer + return {"precision": "A_BF16_W_INT4"} +PY + +llima-compile \ + "${MODEL_PATH}" \ + -c "${CONFIG_PATH}" \ + -o "${OUTPUT_PATH}" \ + --max_num_tokens 8192 \ + --enable_filter_sharing \ + "$@" diff --git a/devkit/models--LiquidAI--LFM2-350M_language_embedding_scales.bin b/devkit/models--LiquidAI--LFM2-350M_language_embedding_scales.bin new file mode 100644 index 0000000000000000000000000000000000000000..d8567528045927cec20291d4c65040734c406ad8 --- /dev/null +++ b/devkit/models--LiquidAI--LFM2-350M_language_embedding_scales.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e215f126992bf8ad3972d4d68419205419bde819d6ebd47cfcfe7675d3d90f39 +size 131072 diff --git a/devkit/models--LiquidAI--LFM2-350M_language_embeddings.bin b/devkit/models--LiquidAI--LFM2-350M_language_embeddings.bin new file mode 100644 index 0000000000000000000000000000000000000000..e6eb24cd167fd6268f1df7f2342c4a0a572b4cba --- /dev/null +++ b/devkit/models--LiquidAI--LFM2-350M_language_embeddings.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:643dc38326a3e929d26afc9eec0e6fde4dcbcd03adac7573d322d01946923ea3 +size 67108864 diff --git a/devkit/models--LiquidAI--LFM2-350M_language_embeddings.npy b/devkit/models--LiquidAI--LFM2-350M_language_embeddings.npy deleted file mode 100644 index a0278700dc65843336207b35fa161a4cec97fc9d..0000000000000000000000000000000000000000 --- a/devkit/models--LiquidAI--LFM2-350M_language_embeddings.npy +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:cdd297bcff8390d6f8504565c0c2fb174de3b0cf7c7586f9793e95632fbe22c5 -size 134217856 diff --git a/devkit/precision.json b/devkit/precision.json index 775049888b4c24c158f02875c2df5a8945f3791b..1e0977f3fbfdb9ae51fd7e645e314c8b41716910 100644 --- a/devkit/precision.json +++ b/devkit/precision.json @@ -2,7 +2,7 @@ { "part": "group_conv", "idx": 0, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -12,7 +12,7 @@ { "part": "group_conv", "idx": 1, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -22,12 +22,12 @@ { "part": "group_pre", "idx": 2, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "group_post", "idx": 2, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_pre", @@ -42,7 +42,7 @@ { "part": "group_conv", "idx": 3, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -52,7 +52,7 @@ { "part": "group_conv", "idx": 4, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -62,12 +62,12 @@ { "part": "group_pre", "idx": 5, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "group_post", "idx": 5, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_pre", @@ -82,7 +82,7 @@ { "part": "group_conv", "idx": 6, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -92,7 +92,7 @@ { "part": "group_conv", "idx": 7, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -102,12 +102,12 @@ { "part": "group_pre", "idx": 8, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "group_post", "idx": 8, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_pre", @@ -122,7 +122,7 @@ { "part": "group_conv", "idx": 9, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -132,12 +132,12 @@ { "part": "group_pre", "idx": 10, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "group_post", "idx": 10, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_pre", @@ -152,7 +152,7 @@ { "part": "group_conv", "idx": 11, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -162,12 +162,12 @@ { "part": "group_pre", "idx": 12, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "group_post", "idx": 12, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_pre", @@ -182,7 +182,7 @@ { "part": "group_conv", "idx": 13, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -192,12 +192,12 @@ { "part": "group_pre", "idx": 14, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "group_post", "idx": 14, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_pre", @@ -212,7 +212,7 @@ { "part": "group_conv", "idx": 15, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "single_conv", @@ -222,32 +222,112 @@ { "part": "group_cache", "idx": 0, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 128, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 256, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 384, + "precision": "A_BF16_W_INT4" }, { "part": "group_cache", - "idx": 320, - "precision": "A_BF16_W_INT8" + "idx": 512, + "precision": "A_BF16_W_INT4" }, { "part": "group_cache", "idx": 640, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" }, { "part": "group_cache", - "idx": 960, - "precision": "A_BF16_W_INT8" + "idx": 768, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 896, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 1024, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 1152, + "precision": "A_BF16_W_INT4" }, { "part": "group_cache", "idx": 1280, - "precision": "A_BF16_W_INT8" + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 1408, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 1536, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 1664, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 1792, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 1920, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 2944, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 3968, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 4992, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 6016, + "precision": "A_BF16_W_INT4" }, { "part": "group_cache", - "idx": 1600, - "precision": "A_BF16_W_INT8" + "idx": 7040, + "precision": "A_BF16_W_INT4" + }, + { + "part": "group_cache", + "idx": 8064, + "precision": "A_BF16_W_INT4" }, { "part": "single_cache", @@ -329,6 +409,36 @@ "idx": 2047, "precision": "A_BF16_W_INT4" }, + { + "part": "single_cache", + "idx": 3071, + "precision": "A_BF16_W_INT4" + }, + { + "part": "single_cache", + "idx": 4095, + "precision": "A_BF16_W_INT4" + }, + { + "part": "single_cache", + "idx": 5119, + "precision": "A_BF16_W_INT4" + }, + { + "part": "single_cache", + "idx": 6143, + "precision": "A_BF16_W_INT4" + }, + { + "part": "single_cache", + "idx": 7167, + "precision": "A_BF16_W_INT4" + }, + { + "part": "single_cache", + "idx": 8191, + "precision": "A_BF16_W_INT4" + }, { "part": "conv_post_final", "idx": 15, diff --git a/devkit/vlm_config.json b/devkit/vlm_config.json index 8b6695ae7468b82b4cae54e2bc172a45e4d50e79..ddda77706c6a4725a3eea2a2d5ba5967bd23efb6 100644 --- a/devkit/vlm_config.json +++ b/devkit/vlm_config.json @@ -20,6 +20,7 @@ "low_freq_factor": 0, "high_freq_factor": 0, "original_max_position_embeddings": 0, + "attention_factor": null, "long_factor": null, "short_factor": null, "rope_type": "default", @@ -83,22 +84,81 @@ "pipeline_cfg": { "system_prompt": null, "chat_template": null, - "max_num_tokens": 2048, - "input_token_group_size": 320, + "max_num_tokens": 8192, + "input_token_group_size": 128, "input_token_group_offsets": [ 0, - 320, + 128, + 256, + 384, + 512, 640, - 960, + 768, + 896, + 1024, + 1152, 1280, - 1600 + 1408, + 1536, + 1664, + 1792, + 1920, + 2048, + 2176, + 2304, + 2432, + 2560, + 2688, + 2816, + 2944, + 3072, + 3200, + 3328, + 3456, + 3584, + 3712, + 3840, + 3968, + 4096, + 4224, + 4352, + 4480, + 4608, + 4736, + 4864, + 4992, + 5120, + 5248, + 5376, + 5504, + 5632, + 5760, + 5888, + 6016, + 6144, + 6272, + 6400, + 6528, + 6656, + 6784, + 6912, + 7040, + 7168, + 7296, + 7424, + 7552, + 7680, + 7808, + 7936, + 8064 ], "future_token_mask_size": 128, + "long_context_future_token_mask_size": 1024, "return_logits": false, - "use_strided_kv_cache": true, - "enable_filter_sharing": false, - "quantize_embeddings": false, + "enable_filter_sharing": true, + "quantize_embeddings": true, + "quantize_kv_cache": true, "split_mlp": true }, "language_model_name": "models--LiquidAI--LFM2-350M_language" -} +} \ No newline at end of file diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token0_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token0_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..d4dcd317e0a511fae530b73b876e1ac73c3f1a5e --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token0_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2000664e9de053efb9879cbe7117995c93956999ba5742a2a7e701ec95a5952d +size 1315808 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1024_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1024_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..6cd0eb4e11b46940d516215a0f3b73b86eac4d4b --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1024_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:137617f0774f4dab6f8dd9a990eb51974d1d63fdcd9eec85abb4ee652dfebe9b +size 2873296 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1152_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1152_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..63da8e46228040ad6cbdfcf6bbade9527a44a35c --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1152_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:36d1e5878dc6b1eae2567bc0d014e0f3a00cfb04ca5f383a8639ad73cc876e85 +size 3726104 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1280_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1280_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..6d1ba14c3762a072f1be9cd2fb3c15a3aa030d0d --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1280_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ad89e2e5b06e6c5f17d93cc25700a51f6458fcd14fafde0d4800c2b86e62279f +size 3738168 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token128_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token128_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..bd7805aafbadc1ab99e38f73c9ba0d397fc7e127 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token128_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:676ffcdc3f8e6eeafd75efe23ee27f68249dd51df2a4776126b6c1ebf4e3a356 +size 1260440 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1408_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1408_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..751655bd062103582faa629524b5b7bcec1eca8a --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1408_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:466e27e08cf91f5a236f539a6203aaf840bd28ac8fc2245f1a311c29a33f9075 +size 5490264 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1536_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1536_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..b8865b31695a0a7099eb2ab8ab8aa000a8f5b196 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1536_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fbc94b2156085dcbb3352aa96fb30cdcbeed1c98e5143a85f19ba98e8be36802 +size 4519552 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1664_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1664_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..1c62ed6c508af33f0f90948c22a6da73fcad616d --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1664_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f0d1f94cf22e220291ce5612385615f452378c926d7deddc81f298509575ec8 +size 4939680 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1792_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1792_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..16347a1e48803afd8782f2f69ac70c7cf7d953ce --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1792_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a0d19d1ce80623ba06fcf055deef3f8dff7e6ab0ea4a7706240b3a8250ad2fe4 +size 4558264 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1920_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1920_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..ced4c182f2cfed248210ab6da94b7d1b77c948a8 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token1920_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:00922f7f12f13408a8504ea1e48140aa95e27fbc64fa3b614d28e6df19393efd +size 4979120 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token256_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token256_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..0a376a0aa310a7d53f82d10bc73b4d48688a013a --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token256_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3cb20dd58f4c9fd4663dc323646def04f6edee19310eaed9d6263a678b66811 +size 1462760 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token2944_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token2944_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..07f50fc34ee28703888d7f07962096eadd5fae35 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token2944_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b994fb8159f5bb4271d01ca46b192a1cae663afa523ac944ced6a2f2672af470 +size 9722584 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token384_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token384_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..40a0e696c241ab8afceadd6df1a89486e76fabc7 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token384_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c5f30653a975c6db6d084bc18cd546f5e89e1e81a26684805562e0f88dfdb46 +size 1632896 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token3968_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token3968_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..d7bc8bf8e626491fd0978bdccac72c2fbeab58ff --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token3968_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cff72489fdb4508cab4f4a769d8c6dfd853a9203276b67ec1f8cca24d9040fee +size 11044112 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token4992_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token4992_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..fbb797542f41a765ca861cdd59ebfda63dd02e76 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token4992_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f8f70ac0b923362f6fed24210776c6e9a14ba9b26ddd8a7f679ca1a2dc921c7c +size 21420168 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token512_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token512_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..942e8d0551f7a149b7a445591d3f9a588d1c90ee --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token512_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8293cfc669a207f46f7bac95f3c6614f00fd5506f4f5b52cb979eee440b2b71d +size 1630136 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token6016_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token6016_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..7a64d498dbf1a062503860ac8af9c0cfa33e323d --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token6016_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:19de5efc8c9699a7cc0f65d8fc6aa2b095aac13eb54506f8b40043cd0e355b05 +size 26151008 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token640_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token640_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..748903f845a01f4efb1bcd46bb9575e2a9b40e24 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token640_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c05c03970a1d2320504b61366980c56c7f7bc38dac44e8820a93704fb6e526e9 +size 2122144 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token7040_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token7040_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..d524b456675f55ad8769b731c15a51178d1bb2ea --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token7040_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:606177940e9dcd98bbcae23059a45523b9fd56c96523082dd06f951a23c24fdc +size 39042256 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token768_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token768_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..254d01ae6d5bd2d87734226fe717a421981351d4 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token768_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a4f8c524f44a879028085c5f98929760587a04893d5b46747d44e7f57eda7650 +size 2149096 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token8064_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token8064_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..e26bb19e406eaf72dfcd209c3e22e9d1b9125cd4 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token8064_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:801f9608972332a2ba448b19bc7e6792ba4a6e22d01c3a51959ddcf3b87bf14d +size 42826192 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token896_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token896_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..5e629ff927583a696c911825054d5bf64efe8cce --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_cache_token896_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59c7fa375af7a760ee15c0ad7414c55013249654305185d03b4d46bb3758f840 +size 2182032 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer0_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer0_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..07ee3f7b153eb8222538118cae70cbd5dc5dc863 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer0_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c96e229d218d49b7dd85807ce1d6263904974184fd743cfebbdc184d01128301 +size 19566408 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer11_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer11_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..050a7560b74a0a4cc4be25f1ed652c0c971e9e25 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer11_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bdc41d9dc668b4d9105b576ec79a27f77303f569cc41a43c343a1377ff3225f6 +size 19760680 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer13_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer13_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..8556a758bf062404d1fc22a7a0a47b1e4e295c73 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer13_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aedca1ce78af5aea5b6e6c412655babc98e928b2449840aaefcb0554c6e8ac28 +size 19760680 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer15_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer15_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..aaafe947343c631b2347c4bc0dcdee90c3f350a4 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer15_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a3e00fc249ef03024062c8862df49081c393d7be036a6fefec9fec96f64ae23 +size 4907464 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer1_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer1_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..66858195ac6f74fb557d7a9cbf4060d10ba2874f --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer1_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5cec6bfd0e82439be82e387a953ab1158c196b32a213270cb5eb786588f97912 +size 19760680 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer3_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer3_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..3dadecf7b6086ca730090ea1af147b6e0581e10a --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer3_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:02803e194f75de3615063e531f29ab2888583d477e171c707b2d3f884e8d888c +size 19760680 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer4_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer4_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..c1232bbe9f6fa11820124e3d163c2fbc0f0b822b --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer4_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31ab9e7052d7b202456dedf38b29dbad6123631cf29b8db11f1fa24e6e2429db +size 19760680 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer6_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer6_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..eed1276c9a545e8c9f3c23b5968e9babcdc4217f --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer6_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:932e370f3f73b6f691aa37b3b5f8e91e6859709a89071b383851350cb7951e58 +size 19760680 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer7_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer7_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..5142253bf838beb418d26e8fd83307e9c1df196e --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer7_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94b41e7972f9afb6cb918001d8b3a82b7b801cba82caa5f088d3de89ef356a1f +size 19760680 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer9_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer9_conv_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..220e4a7f3a2b5bff2bb9b849990be5beb78b8177 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_layer9_conv_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5885a08d100352e40da7ec1358a2e9ba624a2c8a4054b9ecce3656f877db4948 +size 19760680 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer10_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer10_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..e5538517d208d920e90c1bc1ba33d8f1a94a936d --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer10_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0d69b10d21c017389b397c9e648f02f8d2ef321c0c31f42ba74a0158897a918 +size 16346184 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer12_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer12_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..f3e1dd0b92db9a5f7e89d27f4c4638b41ed8f35d --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer12_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3a854d4d79917fdcf7fe8929a35bf15a5f72f61cb306604663dacae57c636d0b +size 16346184 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer14_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer14_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..624115ec914ed8ce13822f029853d3d6015608f6 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer14_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce2a3f9733e13baeb0e0d4fd5e8b64c719d241fb60ba8368c581f1941d99c641 +size 16346184 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer2_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer2_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..bcb97410f9d280ec1898648ca01017cbcf9129e0 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer2_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8633f2ce7299ff681abe3edd10a92e3d098d76ccb296ef9753673aaa6fe4ae0e +size 16346184 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer5_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer5_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..be56aff90909fb7710a0a47fd98dbf9632a6a90c --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer5_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b2ad53eadd84697a63ebe3bce391f10cc658fae9df758f38ea1bba01bab62a1e +size 16346184 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer8_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer8_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..c5f9a11999cf5781f4092bcd17705a6a73f7c05e --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_post_layer8_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef76e514f00a2622ba669b4a5a5c462cb032d5b9c21c928e873b78ec7ea3fc1f +size 16346184 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer10_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer10_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..df7eb0d6c33875ee84331de7abd1c45ea55006b6 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer10_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9260aae766ad7b5e3692a208f2b56036134f0f6fbc0da7f0f06b66f5df160d90 +size 6402752 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer12_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer12_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..3bf406b82f17a8e33f9fdd99ba760f3c5113c7d3 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer12_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aab45c7b98ab3fc8f3b519e0e8911331c0235205183f1cb44c51fbbcd62a25d6 +size 6402752 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer14_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer14_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..244e00d558eaa6c2810ddf524711e9f909fb9905 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer14_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5b23a9c57c8e589a719d2e21f876eb68939d6954b82bcaed4108b613dc588f7 +size 6402752 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer2_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer2_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..23ea59c1118ee9ca91f6109b1fa234d347140d8c --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer2_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab5680239d823dfd22dca9a5bd11d2eecb769527b07ae7332b1fd796076bc636 +size 6402752 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer5_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer5_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..4216bc07ca32d15740a7a92fff4cd5946c020e8b --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer5_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6cb7949e450fb57f4a09e3c331c242545df0b7ec2cd2b157d4c1c35a104af2b5 +size 6402752 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer8_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer8_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..7b9c987359bd10f374b6b44bca3fafc0c32ee06d --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n128_pre_layer8_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a25798ee057449b3e1e70b9e4c2f302e8db3c4e51e3e7b081e29a4f276f2b3a1 +size 6402752 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1023_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1023_stage1_mla.elf index bbf17c58fde91e14379ce438ed43c29af2c23ab9..ac0d48164352792ae1b388f3a85ce4eef50394fe 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1023_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1023_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:5d7d6ac5070ac345938bc3773a753a6a8706dcb8d7e0cfd1bd8fe97737564baa -size 928752 +oid sha256:6aee101758cb67301e05efd4cea4bc514138b8134c355e3b64b07029dcb0aa74 +size 982304 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1151_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1151_stage1_mla.elf index fd09a6cffeddb54c73ef677e0bdba57f926e5614..e3394c322f27887d041395da7c0908ddc6785dc9 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1151_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1151_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:fd5f5f9e82def6902322c6b70a1df50e1e17a6b877492b5fe0b7618e4cfaee3b -size 777960 +oid sha256:0155cbddd1ba52a45ae5592eeb3eb247420f3733fa9ca0f17cffbccf2932615f +size 864064 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1279_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1279_stage1_mla.elf index 26037add1417c5687baa2d66bc17d612901fd727..8125f4b4611b381859f8fd42a93af55475ee2136 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1279_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1279_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:e95fa1631007cb83d80246aaa54b1004b217353251953a9fd11b7bea09c7e8cb -size 964080 +oid sha256:af50cf8659a0ac2acb1f12498e06185d536e494201c81236354562805e08a4be +size 1045632 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token127_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token127_stage1_mla.elf index 6424ba98d012645341dc4c92d835fc1155a69433..d79efbddc35c583ee3e93ab6fcf8de82afe1ecc3 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token127_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token127_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:372223519453919372610e864158e636a274fea9b585efe43e9d7755ef5ef85f -size 676872 +oid sha256:aa55ddf747263c0af2498d35a56e3cd7d3ce93703e6576042050f1104d1809fd +size 749920 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1407_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1407_stage1_mla.elf index eb7590cd55d0a3d241a644a3691d11bfa6c33bf9..44b80ab44ef96989bf93691f75e2cf4a4c6ae137 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1407_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1407_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:0b52cba6962f309c05a9ccae40c0f846f8bb05d9a383a7d58a216b702dfebdb2 -size 802248 +oid sha256:686563b9dab6a60d53071351d20140272cdd72ca7c70f9b2c3ce7b474c462be6 +size 882336 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1535_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1535_stage1_mla.elf index 93b4d36a6e9cdd3a58720e4df94b9396f8568e67..0b83b57f4c34e8bcb2f3d9ff6bfa23a949a0cec9 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1535_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1535_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:37619ce751feff072f1b6b226ec363a7e5b8076b1ba4224d603fee77a097fc1b -size 980016 +oid sha256:16839ea96265f0dfabe796948f4d7dbedb1632dc088e30b93b6c4353118425b4 +size 1061904 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1663_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1663_stage1_mla.elf index 531d06d8ecbf715bbb8855509da73be0d62a797e..e438d471c2f9f801b68af23bf62f927ad31c8a29 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1663_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1663_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:122288dce8b4fdee73f796f816d1c1cf21d3fc4222559258b77d495fff223e1c -size 817680 +oid sha256:c3f78cf947615fcf299d37c06541667fb96526614caeb196efd349d139c94197 +size 900048 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1791_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1791_stage1_mla.elf index dad00fd4b9ec108f54e1773827f43a232c2e7361..4bb156882df65a54b573837c0de1a3a145cc83a3 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1791_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1791_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:fc1651a7b32ca3dc5f9bc5273c3db0aa3269d68b95220c4a8693af5d28776451 -size 983648 +oid sha256:21eeb0a0fe4974891627e556ef16cf179433f44227b951f62be1b630dc364afe +size 1067288 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1919_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1919_stage1_mla.elf index 99b9e00a7995c8b27d7c3448076b22a21f6ad71f..6c3bcc15a1c97bb1439122a6a625855937335892 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1919_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token1919_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:1ab2596e9c3eac49d2d5e46bec7aea4fe316c35468ee382b7e9f5df7efd01c24 -size 829136 +oid sha256:8d6af2967b5750a7ea7d60e5bfe25f854b5f0a6ea2701b755ed7e7cae05b42d5 +size 911328 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token2047_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token2047_stage1_mla.elf index e4aa4792f0d1edbd7730cfb4c0a27062482f7d19..a6025dfd5995139210baac92854820a47e869759 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token2047_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token2047_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:355e5f63af5ff649247848db291cce3c192bf94a1ac0b4457e69c34852a04356 -size 1005120 +oid sha256:3b9197f5fc0df5b8bed6c2a3a1f354c1e4be229e59532ad952bd1294433ad280 +size 1089824 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token255_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token255_stage1_mla.elf index 68438e02e32468a6721d31a15d4189915d264abf..4d5e60457aa592ee5491a63aaf0b6aa0b99027a9 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token255_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token255_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:74d1e2b8cd514be262f61cb0f16e28c254b63d48d39c65754bfd7f6a94825ddd -size 770456 +oid sha256:5005f279669b561c6692d80d8a7ec60769a1bfbf2f7fa07133710136fc5d2922 +size 761088 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token3071_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token3071_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..a10f533f8e6d96f23ea26876c25bb296e758b386 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token3071_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b5a777e20c1089214126dd0ee0e9b06bb10dafc617346866032adfd1f3dd4d87 +size 1751848 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token383_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token383_stage1_mla.elf index f97e916d9f36c8860130c6272a8fff0005ac50d9..a0c2eb7f3bd7005c536462d96d78187b5ff20e97 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token383_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token383_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:1d15ec46827d662da6ec9f2c8f87251af3bf764ea5e2c24f7e56a442ca00b5da -size 712800 +oid sha256:55604ad7793d50cd95721689ca32cc0f96e59132d1cd3c2319e890e00a3941e7 +size 786600 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token4095_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token4095_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..149e3d52dd6a5f836d0f92892199b3f5b6c900f0 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token4095_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a193e58c6054e9f8bfeb431dc1f19352dabe830ec3d74066ecc9a94477921f90 +size 2046696 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token5119_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token5119_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..53910292031a9f37011a887e77a3f5ddb55ae5d1 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token5119_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5c33c14c908c8fc218ac46615b819b9b2a4cad85f57e594fdc03c60c5d408a9 +size 2474648 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token511_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token511_stage1_mla.elf index b7d1630e95fe09c3dc2a624479a3ed711d7508bc..1188865abe6959213a6f95fd408edf78db243c31 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token511_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token511_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:a62d011bb300ff3fa0dcb004b5918d8d2f126a4ba680a94416b93f73ef00bd3c -size 894184 +oid sha256:66f7a9f4938a3bf02ce59ec9c6b65d4eaa3e6165f5fb7317dd1e026e79bdb94c +size 967528 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token6143_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token6143_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..8a3bbf7b3898c329f1ac0b0159e35e261a008fa9 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token6143_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77620822df4b39525c5ad33b1c525b5a8502aac22da9682fbe79166c74ccf254 +size 2776616 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token639_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token639_stage1_mla.elf index 839ea2302c2f3f9d37f4d6c723cbe8bee8bcda9a..e164e236ccf60a205788b964e18ec5d5ad5dc124 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token639_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token639_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:2c78079d0e9de248007bf6d550a2daa234a398dc4e6c3a2de0bfc9b847531867 -size 735264 +oid sha256:768c47a69100af4bae826b44ddf85d1e5424160359124edf10de718d0919934b +size 814232 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token7167_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token7167_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..eba71e77dd21b26ddc3adf542315b06d36f96aa8 --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token7167_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:63f760308cb2aa4d31451e3c7e9c4928eac8bc0a02e361c7c735e3e70dc305d8 +size 4011368 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token767_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token767_stage1_mla.elf index 33047843e548af1df2a615c8f58d493ae52e0f88..8d2cb48f36a86ff11aa5229e31a7092dad9c9373 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token767_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token767_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:635bf18a115c7982d35b7d42cf700ea208782032584d2bb35f5f348bbd915534 -size 894064 +oid sha256:b68895f1a2d350c5567dcf818a16c0701e4d9062a3903b2d917a6953ed99b50b +size 980280 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token8191_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token8191_stage1_mla.elf new file mode 100644 index 0000000000000000000000000000000000000000..af29c818d193bfc376af8db050aa827a1921426b --- /dev/null +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token8191_stage1_mla.elf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14d68a50783cc3e20f86596a587de00303e1e403b3247293c5f91453a1f685ee +size 4241208 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token895_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token895_stage1_mla.elf index ae942da541c13165b70fdd2b91d076ec5d188340..ae8223d14a90cdf334fdb7ee3f75ae7bac347a98 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token895_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_cache_token895_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:e415acb0fcb0c2a0ae62acdc6d6f83c9a98dde56d86d10d8ad13be4b4acbb128 -size 745600 +oid sha256:50e4b7748e6a47a7904ab712016d106977cd7a007e4ae6a8bb74683757778d50 +size 832688 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer0_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer0_conv_stage1_mla.elf index 40b2a8b3b54bec27af01d463fa98cf03de53d745..e15d2ffdfc843cf93157944a6891d5bccad3a809 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer0_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer0_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:f1ac34cc39da1bb41c2968bf888fa8988d809bb5ba74f392f70ff09305c175ec -size 10948888 +oid sha256:a01ae6ca33457187aed0c9cadff10691f71074e93d4b7147ab11b305dbc2ffe9 +size 11649384 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer11_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer11_conv_stage1_mla.elf index f8a84c47fdba6b470c404ae8ad7ac0f01a206668..f658112e7489269fbb1576941836e5cfd53dc0b3 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer11_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer11_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:908ae8f18ce00c34937ca7e545451e5c0aa359ae46f4df28718863fe6164d268 -size 10893352 +oid sha256:6ad2934519181b863ea94c4f596b0d3ebec5c09c3d2b94cfea241236b73115cf +size 11667544 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer13_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer13_conv_stage1_mla.elf index 3423a68a4d2064afd36e7b388a7ca826a1b530f9..00b2311fc9ef406421ad41e9ffacc307d9f16d99 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer13_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer13_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:f37843ca96fd829a8bd6391a7d9063b0b2f557851de7dce6f8aa2f3b80e9a757 -size 10925072 +oid sha256:f6a203dd1c28e1c21232f3f1a57a79788aa70c2f413fb564d3ff5b9270043cd2 +size 11667544 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer15_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer15_conv_stage1_mla.elf index 4ef7d4c472524a7c7b74648450873ccf0c422f4b..2e53b28fbb9e14e8b0f063cee0544f66f54fdda1 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer15_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer15_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:8fe37d2f2a4ef8179acee9d80ad00cf73a900fefdc20649f95ce2b05a4d34965 -size 2713856 +oid sha256:201ae98903d2befb405518566c8a8ecb7e9ba408f63a1aa870081da1b43279b0 +size 2879624 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer1_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer1_conv_stage1_mla.elf index 59229a9057f375685eeef9364ebaeda60370ed98..5a11dd9d37f83421c42e0bef4996a461f0e42d87 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer1_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer1_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:12739177ada1958a1d6c2b5d36dc828088b53de6019ec7c92eee7e06d62756e7 -size 10942904 +oid sha256:2e3bab86a946ceb36b9a42d291b27718b534fc8ad8a320b190e764239428c41c +size 11667544 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer3_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer3_conv_stage1_mla.elf index 3267e23fe9e2f07310d3ef81ccf62803d29ee7dc..a051e354c6aa9cb5d36151153f8895b150850377 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer3_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer3_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:0ebc7f2538a57f83bd8e1029517f382421aa42e66ec0992db2828ddd3498de54 -size 10932248 +oid sha256:a3b74e8d90ebd9e392cfb6dbcbb4e07a287432a9f4a439220a05e9f2eceadfb4 +size 11667544 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer4_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer4_conv_stage1_mla.elf index b85670d906059452e04600edf5c32d49ba10917a..509db15d5599de72b6ddf454f851bd887e10f4b3 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer4_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer4_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:9444b7c1ef3d67aecf08b36d254c7e2c3bf632a1e434c4fc241801cf0d1b733a -size 10873288 +oid sha256:02ea08ca5dd922937da189bdebd13d08d0ec93878bd5056f8a7afce5a9faeabf +size 11667544 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer6_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer6_conv_stage1_mla.elf index 21174f9b9bac2a761bfab275039920fa39cf8417..30f3674c59f05b37d5af0e019c49bec4bad02127 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer6_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer6_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:f8b6414f566c2b24063049e73da25ba2c32bac0462b3d3a4cfcbe678dc3568b8 -size 10927112 +oid sha256:a646580166e345cd1e5053b7fb24b95a5a70df73aa9c2f0261522056faa77ab3 +size 11667544 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer7_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer7_conv_stage1_mla.elf index 885a610e536a845714b4871fec85aae719d9829b..50f9d265cfee134e96dd59d0dcba5c0da6f155d4 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer7_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer7_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:652316c25ddda1389c340b538af8a660f5fe9b4b88d4f73f10d06c5dc0f55412 -size 10869192 +oid sha256:73dfb1cc59fb929570ba88ab7a987aadebc49523f7fceac583e6f43c722b3c73 +size 11667544 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer9_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer9_conv_stage1_mla.elf index 1aa2b3955ca7a8dcbeabdca84a0a87b8bc02fa24..719c7c450baf6a0057d6caa4b08fcd3df05d9bcc 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer9_conv_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_layer9_conv_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:76d2b96de75cae88b284ed55198ab3db7a254e454c732fae3bfb7ce3fa75c395 -size 10841928 +oid sha256:3f15f273f3aa608c87b5910288a8f3565da1d7552e9d202634bb750255950fa0 +size 11667544 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer10_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer10_stage1_mla.elf index bb52109c492bbdf229365504bd3d4e30542e526b..1a7306c0711edf59a01160303a1d5738887bec13 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer10_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer10_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:c9d9c8114fc7f6f182ca3a3f0eee0a78ee3f38fe336c3d2687f98038bf6b5d3b -size 8851648 +oid sha256:8ab304500ee9997a24ef4f058090049fc3bd32748641134af2c451c9ce15b2a3 +size 9542864 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer12_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer12_stage1_mla.elf index 7f82b4d284642b5c22c05bbc98d7e0f0abcc1722..7246be15d15eec784be35329b6e5babd1dafc48d 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer12_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer12_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:d08fd0686d201d4b3827d4497b1735dce523a4da2b6ad69e8f90014a8e2ee171 -size 8933376 +oid sha256:66396e4d4bbd7838fc07c92379ed48d04bfa9931b61ce4e24e5bf6436dc52a02 +size 9542864 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer14_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer14_stage1_mla.elf index 7dcde714f49b0d7d3bd518562843497717163488..dea79965bea8c393585afed4682a206b688d4edf 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer14_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer14_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:6d4012b8a88d4f63380cbd7ba4b97cec2cb3aa9ac48ec5f2a5774cf7f6382146 -size 8915216 +oid sha256:12d598da2570b4fde03b94f744fa69b3c17db6a7a674c6791681b32557ba2639 +size 9542864 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer15_conv_final_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer15_conv_final_stage1_mla.elf index bd9ebec71cff208dda0d22332e249b7bf2662055..3426302262eb58d7a42a0a00c68794205c11799a 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer15_conv_final_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer15_conv_final_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:ab4fd96d9c82db957547dbf2193f2daa04defb81eecc9b9867f83a626c6541d6 -size 42972400 +oid sha256:e60b4c731800013cd53080728e4483709fef4427ebd17588750550362d1a01b2 +size 47680312 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer2_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer2_stage1_mla.elf index 094bb8800768013c3db5231909364d98eb00ea7f..498bf155cea992a1fccb80e667a6a78af476c62e 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer2_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer2_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:5cfaee5dbd7ea26f7de1cec84ffc2f46291e231a9e372fb83b90ba18af7acf02 -size 8920616 +oid sha256:fcd36bc5e3a7d6e1f36fd20d20851b841387e862a0198df78ec64f18823ac9b3 +size 9542864 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer5_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer5_stage1_mla.elf index 46a2cc49cc5d14700c609e779386a87fe2b81eea..03120245c21c37551c40f51445907a9631820f50 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer5_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer5_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:62b5dac5acaef15496be41ddffa46b95f27f384397415bbcd7a6328827d90053 -size 8913592 +oid sha256:dddd6301740f7230f951fd7e0a85cfec16f9e85950b14f03c90f82e7caa7a6dd +size 9542864 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer8_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer8_stage1_mla.elf index eba114a8ce3ba9a691b1ed406bf6b34d6016ab57..f0840c3dff7543e1a41aebb1cb5c9e8cf1f5c3f7 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer8_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_post_layer8_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:3a638bf6f84924e17e062f7de2a7105bdaaff26a646675bda09d352015d255dd -size 8911368 +oid sha256:b925dd721f52541e436ece6597c9bc097634e4b2f93dc396b2d7e63cf0b4269d +size 9542864 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer10_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer10_stage1_mla.elf index dc4ec90e552ede593b67e931fea54d837395c7dd..a2b6f4949362d07f809a95a397f259f27153b757 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer10_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer10_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:ef130e63df44b1e7fcebc3195068a16c165d0a5cbd101fc31ae56de0a90a00f0 -size 1761120 +oid sha256:19c0d7d9c2d923f50dd9bedfd1fb00f1aa7b167f743bfaaa75b088642d9b042b +size 1992096 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer12_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer12_stage1_mla.elf index dde18213caf1999490f18ef155eddefadcece7a7..6ffa2b485d09090afa4b113eb482e0cde0b01dba 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer12_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer12_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:07a373f6b6cbbc63af03fd0f7a2b129ab83e34ff53ec25049b3f66765ae7afe9 -size 1761784 +oid sha256:c095a128e2e69f13b6435f75c29d06b948cf451582e7c3408271c8c6b7c6de80 +size 1992096 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer14_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer14_stage1_mla.elf index a238e3f05408a9eaecb024842f5d698fa993803b..49f00c4979c3948dae497ab61fd077631a43cfdb 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer14_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer14_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:f4686fdd520a1d843ef5fa53585ef2b7d0e1b135e9da757bf754546be0047827 -size 1743912 +oid sha256:e798d187c2ead50239c27c46e915a9a79474dcd3ef76940f86d9999e47a45ad7 +size 1992096 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer2_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer2_stage1_mla.elf index 02510b0a091f91c0a77f94247265cf03f061b1b8..40b5179ce8fe673f80119067d42f6e630b66d032 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer2_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer2_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:167cc91894ea7d1aa098402cd343de6fcef3564646bae29d39e96b652ea38930 -size 1771688 +oid sha256:c0ddb8b82d2d8a414b1b6c46d7a845f29d745e4b88a400a88e25a4a6f48947e9 +size 1992096 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer5_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer5_stage1_mla.elf index c3728e7e805779aa351962e0a5f403345b6797a1..cde4077685c5618e18204858b0978e58c6214645 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer5_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer5_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:635f4a9c71a8b3b23f0a30bcb89c12854b6a8af391153e997d9ad18b6931fc13 -size 1769656 +oid sha256:efc278d5d41a2e672b5f3bea0a85bb45c1e1a9b965e90324863343cc9930190b +size 1992096 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer8_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer8_stage1_mla.elf index 2253d77f76b44844ce9d7bfc6443cc9309b27dac..2be7ad8472fa5c068f71a8577db708798e3e7e63 100644 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer8_stage1_mla.elf +++ b/elf_files/models--LiquidAI--LFM2-350M_language_n1_pre_layer8_stage1_mla.elf @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:1a1190af6dfab99e3258b4a006cebef54c3abeab6c29c2f6f9e47f09d6c56bea -size 1774360 +oid sha256:cf3fde98f1a3f32cfb3f19898ce9556329b15fd5916af2131c9d459a6755f82c +size 1992096 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token0_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token0_stage1_mla.elf deleted file mode 100644 index 0ad0c5463f45c6d04e5827f61b402abd45857c4e..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token0_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38f78e0778f876d44ff8b91a464439dd6233fbe73c30e7419e63a23273f670fe -size 2859496 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token1280_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token1280_stage1_mla.elf deleted file mode 100644 index 3e6508827624da4a1c56fddcb43568d4b167d0b0..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token1280_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef2cc909cf5271a5a38f2dac5e62a6735cfec3b4d1b8fff799cad8579136084a -size 9041824 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token1600_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token1600_stage1_mla.elf deleted file mode 100644 index 5f92f8e148393b2baf84a1eb1d5a81deb61f517f..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token1600_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fc68cf26ba70f393f67642563da4421a51d554f416714aedca8747ae87f05d42 -size 8547832 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token320_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token320_stage1_mla.elf deleted file mode 100644 index 1fbcdb1bbcce641dae22fb6a91d2e43e4edb125e..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token320_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da3769d617f8ddef55c2a7ff4d044a11c9ec16d1a987658760be7ff5f6720bbd -size 5222960 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token640_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token640_stage1_mla.elf deleted file mode 100644 index 1f64986347eb950c04235dbbe6cd3de55adde9c8..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token640_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38d90d2c3673bfa285a693287962e22d9739f821030302419d8c26b3a2244eea -size 5484944 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token960_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token960_stage1_mla.elf deleted file mode 100644 index 44758eb3f10994619ecdd433a237cbd4f09a059f..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_cache_token960_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f8f61432fca0babeb9122688d59babf7ed672ed839937e9e8c0f5097921f34d1 -size 8121144 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer0_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer0_conv_stage1_mla.elf deleted file mode 100644 index edb943c2bf71fc553e821f9ee5d3390202546adc..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer0_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ff653da0652627034af89620947991ad9e6e33e5d59451f90b4c3c58ab226b5e -size 28251328 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer11_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer11_conv_stage1_mla.elf deleted file mode 100644 index 3046ff169f8022cd1a5c792ac39e243ce4fad32e..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer11_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccede51432b807e8cd8b12c4b4bc8be951d4e98286afe2d6c21443009624f15e -size 28101296 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer13_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer13_conv_stage1_mla.elf deleted file mode 100644 index 993da6482fecd1190d44beb537c7d4ef2018f047..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer13_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7f3e1a7e53b2d92f41f6421b2ac318d272ab0238ddec1e14580aaad7f34150a5 -size 28195344 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer15_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer15_conv_stage1_mla.elf deleted file mode 100644 index 0c5a959f7a65e37c5465a653d0f7b88315b9cd6f..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer15_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9f88ecb0d51cc36638124f4b25c130ab739d07eaadfb37ad30fd5f3021312579 -size 6533496 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer1_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer1_conv_stage1_mla.elf deleted file mode 100644 index 96ecafe5ccf66697da0a547d3ab4bd496615d8e0..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer1_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8c94653fdaa8d0587d0a342edbafaa53d4eb536c0794563e4a52732f718018cd -size 28250448 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer3_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer3_conv_stage1_mla.elf deleted file mode 100644 index 234a2d8b0f46b0d9462877fcdcc0c802a0003b84..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer3_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7814504b0ee238b1490cf78258c95652f44185073e813b1f6fc42aba32c1b2a1 -size 28222400 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer4_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer4_conv_stage1_mla.elf deleted file mode 100644 index ecdb1a15c189fb94b4eed4e30d983d3d20e3e48d..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer4_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:5b4e95ae4471e2599570b3349205d4e96ce29d40adb559134cf459f2fbd4c0da -size 28065616 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer6_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer6_conv_stage1_mla.elf deleted file mode 100644 index 13b734ffa9d1b42e6d874db558c0894759383c1f..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer6_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9874204f47fc6ce666342a86b41faacdf5d58c2287d63e7f81a6344a3ca74d19 -size 28208080 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer7_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer7_conv_stage1_mla.elf deleted file mode 100644 index cab56d93248d1dfcc38ee45b33f42ebd811e2646..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer7_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:01b795e25c09e7e85dfefa3948fe0361ac3e7688b5cbd4953e86b3bd29da0003 -size 28037104 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer9_conv_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer9_conv_stage1_mla.elf deleted file mode 100644 index 54e43475149c4301df0329ede52374f0dbb98c2b..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_layer9_conv_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a39a68cd0579d8ec01a60b1c50d8c4262b9467ba77be7999b886175b3659cc25 -size 27966784 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer10_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer10_stage1_mla.elf deleted file mode 100644 index 8480919ab56d9bd40bcad4badf1b5b03140e0962..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer10_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f30a7e37a982f6e4d29bfc172aece28efa59734d1141e80fdebd06b1ea3dfedf -size 23588912 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer12_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer12_stage1_mla.elf deleted file mode 100644 index ee9e2528d7e08ab8f132514629254b95916e9e8c..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer12_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:432e429edb4885163c130b6b481aadeab60e9d720c65bd306dec2dfdc6f0c121 -size 23797760 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer14_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer14_stage1_mla.elf deleted file mode 100644 index 80f6079688c4fd8b3be614107b58b77a02a181aa..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer14_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d82d11b7308c811c69b9d71e12a06afa617aa62f58ac5c9702733c8fa7466e85 -size 23742192 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer2_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer2_stage1_mla.elf deleted file mode 100644 index 6420ef5ebfad2fc415e7d116a0d05d0a9102a20b..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer2_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:acfa833b15596039464d13be53401ad27c79fa1ac643891af76d753c6ee9eb2f -size 23763744 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer5_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer5_stage1_mla.elf deleted file mode 100644 index 6c3ec9c66dda2e9ad70259b810d94d5397584129..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer5_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ee329920ac2b131b69ef992653fac0c4dc784173d2c9485381af6c2736f38e80 -size 23746976 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer8_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer8_stage1_mla.elf deleted file mode 100644 index a1262f6f1a5476673b8e5fe4a577699aabb034a0..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_post_layer8_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:dcef9e410513628b8344cb20a0c6f841ef4ec44e6af25d798780c980adfd149e -size 23738784 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer10_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer10_stage1_mla.elf deleted file mode 100644 index d4c1835a8d59fb601d65aa71376866c5e6e609b1..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer10_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7031d90342f26178ce40d39749f00ede875e2f167967f7d81ac747a7cf4a8378 -size 6764824 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer12_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer12_stage1_mla.elf deleted file mode 100644 index c0cdcd26297e6724ddc4e7a70ecae9cf3ba5d157..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer12_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:5d8baa62fb123c486b229b8f060937b2c22e56b0c853eecb2f7e9d344f359fde -size 6771720 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer14_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer14_stage1_mla.elf deleted file mode 100644 index 1eabbb198d949cc19703c0ca6ccb7b84b7c7c2b4..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer14_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7fa7a1ad4149b0c3adfb30697d4601ff54877440dd7b7bf49aa7a778046324ea -size 6717544 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer2_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer2_stage1_mla.elf deleted file mode 100644 index 8c8cb0598f1250fa07d1b662182e544b8c50e3c6..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer2_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:17d120efe74f295dd39aaf5415c4e2a05a2757c9898d3332e7018a0ff67823d0 -size 6786072 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer5_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer5_stage1_mla.elf deleted file mode 100644 index e32f35b29bb4ac62cd0088d2a765d844ad090f99..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer5_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c75388f5e12bc77e7afe915dcc87e9f73ae243c3f4c0572bcbaf2f95732ada46 -size 6783736 diff --git a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer8_stage1_mla.elf b/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer8_stage1_mla.elf deleted file mode 100644 index 67225d64d35ff0eca3a0c320ca3addfb21f71c13..0000000000000000000000000000000000000000 --- a/elf_files/models--LiquidAI--LFM2-350M_language_n320_pre_layer8_stage1_mla.elf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e37cd0d13544f2086bfba4976fa56f9e839aff2574f4eb324968803a8765b0be -size 6792376