# Copyright (c) Meta Platforms, Inc. and affiliates. # All rights reserved. # # This source code is licensed under the BSD 3-Clause license found in the # LICENSE file in the root directory of this source tree. # For llama3.1-8B python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.1-8B --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.1-8B --quantization float8dq-row --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.1-8B --quantization float8dq-tensor --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.1-8B --quantization float8wo --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.1-8B --quantization int4wo-128 --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.1-8B --quantization int8wo --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.1-8B --quantization int8dq --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.1-8B --quantization gemlitewo-128-4 --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.1-8B --quantization gemlitewo-128-8 --tasks wikitext hellaswag # For llama3.2-3B python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.2-3B --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.2-3B --quantization float8dq-row --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.2-3B --quantization float8dq-tensor --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.2-3B --quantization float8wo --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.2-3B --quantization int4wo-128 --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.2-3B --quantization int8wo --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.2-3B --quantization int8dq --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.2-3B --quantization gemlitewo-128-4 --tasks wikitext hellaswag python benchmarks/_models/eval_hf_models.py --model_id meta-llama/Llama-3.2-3B --quantization gemlitewo-128-8 --tasks wikitext hellaswag