save a TSV

latest_results.tsv

@@ -0,0 +1,18 @@
+	eval_name	Precision	Type	T	Weight type	Architecture	Model	Average ⬆️	Hub License	#Params (B)	Model sha	Hub ❤️	Available on the hub	Code	Data	AMI 2020 Agg	AMI 2020 Miso	ARC-C	Belebele	GeNTE Neutralizing	HaSpeeDe2 HS	HaSpeeDe2 Stereo	HateCheck	HONEST	IronITA Irony	IronITA Sarcasm	ItaCoLA	News Sum	SENTIPOLC	SQuAD it	TruthfulQA	XCOPA
+6	meta-llama_Meta-Llama-3-8B-Instruct_bfloat16	bfloat16	fine-tuned	🔶	Delta	LlamaForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">meta-llama/Meta-Llama-3-8B-Instruct</a>"	58.748316329193344	Meta Llama 3 Community License	8.0		0	True	🙈	🙈	55.37407439022941	71.59839304531086	42.57679180887372	82.0	32.48322147651007	70.53457622533335	63.09031737569537	81.04353954390334	100.0	68.90825671526659	50.63388859343638	0.2575796842123843	35.87793977181792	44.40535171743039	76.4493013414765	51.688145906790595	71.8
+5	mistralai_Mistral-7B-Instruct-v0.2_bfloat16	bfloat16	fine-tuned	🔶	Delta	MistralForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">mistralai/Mistral-7B-Instruct-v0.2</a>"	57.34121935588332	Apache 2.0	7.0		0	True	🙈	🙈	59.26344649551212	67.03848859411114	44.36860068259386	67.55555555555556	29.12751677852349	70.94842426874283	66.92711073442074	77.91591984780963	100.0	60.340552982611825	52.5864148320762	0.2650337064892725	36.39365330456299	50.86004322897759	67.76589485305061	59.24407318497844	64.2
+7	meta-llama_Meta-Llama-3-8B_bfloat16	bfloat16	pretrained	🟢	Original	LlamaForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/meta-llama/Meta-Llama-3-8B"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">meta-llama/Meta-Llama-3-8B</a>"	56.06703915537942	Meta Llama 3 Community License	8.0		0	True	🙈	🙈	59.167006941608825	65.2988113338495	40.44368600682594	75.88888888888889	29.664429530201343	66.34318803515383	59.665954331496216	80.46901075930542	100.0	55.417040602648825	56.72119925007975	0.27369249994767686	32.8415569535643	41.65027333775969	76.0261495015472	42.068777668572736	71.2
+11	mii-community_zefiro-7b-dpo-ITA_bfloat16	bfloat16	fine-tuned	🔶	Adapter	MistralForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/mii-community/zefiro-7b-dpo-ITA"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">mii-community/zefiro-7b-dpo-ITA</a>"	55.965402990833326	Apache 2.0	7.0		0	True	🙈	👁	58.82330921555731	65.29219074291716	44.19795221843004	66.11111111111111	29.395973154362416	66.42034413085725	62.04374417717792	82.92405607588724	100.0	59.58686440677966	54.61088096497907	0.15622781170005148	35.73603929970904	40.115316478607326	74.25556784297711	43.342273213113806	68.4
+12	mii-community_zefiro-7b-sft-ITA_bfloat16	bfloat16	fine-tuned	🔶	Adapter	MistralForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/mii-community/zefiro-7b-sft-ITA"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">mii-community/zefiro-7b-sft-ITA</a>"	55.49594502634554	Apache 2.0	7.0		0	True	🙈	👁	59.05979939301703	65.1057440915327	42.491467576791806	68.11111111111111	26.845637583892618	66.26712374430319	62.8192509112937	82.66496052951742	100.0	52.30611640858258	51.83751520046043	0.1357069141230042	34.79253286178762	46.95941666858784	74.51464966490876	42.52003278796419	67.0
+10	mii-community_zefiro-7b-base-ITA_bfloat16	bfloat16	fine-tuned	🔶	Delta	MistralForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/mii-community/zefiro-7b-base-ITA"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">mii-community/zefiro-7b-base-ITA</a>"	55.00699465875708	Apache 2.0	7.0		0	True	🙈	👁	58.26528760660498	64.28826512391971	41.04095563139932	58.77777777777777	27.651006711409398	63.41040091554036	60.20187319698322	83.36773972540995	100.0	59.61581980369971	57.22956187895212	0.21630746589700614	34.14146108746794	38.60348969137316	75.51969438076942	46.18926820166605	66.60000000000001
+4	mistralai_Mistral-7B-v0.1_bfloat16	bfloat16	pretrained	🟢	Original	MistralForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/mistralai/Mistral-7B-v0.1"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">mistralai/Mistral-7B-v0.1</a>"	54.53886941414389	Apache 2.0	7.0		0	True	🙈	🙈	57.328824884373255	65.894796072133	41.12627986348123	65.55555555555556	29.395973154362416	60.74292449685459	58.40138983607699	81.20893551611952	100.0	55.21599398531273	56.0842910054169	0.21650562273812077	33.95578203972551	38.248077168561004	74.9929389324236	43.192511907311555	65.60000000000001
+13	swap-uniba_LLaMAntino-2-chat-13b-hf-ITA_bfloat16	bfloat16	fine-tuned	🔶	Adapter	LlamaForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/swap-uniba/LLaMAntino-2-chat-13b-hf-ITA"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">swap-uniba/LLaMAntino-2-chat-13b-hf-ITA</a>"	54.11970329791716	Llama 2 Community License	13.0		0	True	🙈	👁	61.10842468417231	65.37114603439397	39.16382252559727	60.22222222222222	25.369127516778523	69.19701593869706	58.47240303675274	81.91924285348375	100.0	60.50989600805099	52.82407691311843	0.14705407414144434	23.961182038838874	33.936431374370564	72.99623578596571	44.43667505800782	70.39999999999999
+9	meta-llama_Llama-2-13b-hf_bfloat16	bfloat16	pretrained	🟢	Original	LlamaForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/meta-llama/Llama-2-13b-hf"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">meta-llama/Llama-2-13b-hf</a>"	51.53259991565551	Llama 2 Community License	13.0		0	True	🙈	🙈	55.5211353099392	59.73745072519405	39.67576791808874	49.77777777777778	24.295302013422816	56.705263521819575	55.58451703385505	75.35374357525852	100.0	49.6392951529161	51.32659342493766	0.15611794645515564	34.99992804182015	35.591589638147205	75.37248669035945	42.917229796152284	69.39999999999999
+15	swap-uniba_LLaMAntino-2-13b-hf-ITA_bfloat16	bfloat16	fine-tuned	🔶	Adapter	LlamaForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/swap-uniba/LLaMAntino-2-13b-hf-ITA"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">swap-uniba/LLaMAntino-2-13b-hf-ITA</a>"	51.398892791161465	Llama 2 Community License	13.0		0	True	🙈	👁	56.91493042765838	60.79803569083185	38.395904436860064	52.22222222222223	24.563758389261743	59.591680814940574	53.72166074176572	68.63908831908832	100.0	53.8835564536499	55.220925077582386	0.24355772539252643	23.46778181911886	37.868993755237724	74.32140387879224	42.12767769734223	71.8
+3	g8a9_tweety-mistral-7b_bfloat16	bfloat16	fine-tuned	🔶	Delta	MistralForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/g8a9/tweety-mistral-7b"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">g8a9/tweety-mistral-7b</a>"	48.30841019247476	Apache 2.0	7.0		0	True	👁	👁	51.45449792748049	56.83712780075503	38.310580204778155	49.77777777777778	26.308724832214764	56.756734367216744	54.259763500716296	64.359704127708	100.0	48.96104026840812	49.87333014539054	0.12625704978630167	18.72596344839197	30.051768572855263	64.28422203983018	37.75548120876116	73.4
+8	meta-llama_Llama-2-7b-hf_bfloat16	bfloat16	pretrained	🟢	Original	LlamaForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/meta-llama/Llama-2-7b-hf"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">meta-llama/Llama-2-7b-hf</a>"	47.88766168947672	Llama 2 Community License	7.0		0	True	🙈	🙈	50.167656275074535	58.36785332162082	34.8976109215017	36.0	24.832214765100673	51.08771929824562	54.388067109409945	68.27095354111434	100.0	47.98695094164673	52.28499188648629	0.1232306318769991	33.83386905556545	28.13533353128773	68.54722302033736	39.16657442183617	66.0
+14	swap-uniba_LLaMAntino-2-7b-hf-ITA_bfloat16	bfloat16	fine-tuned	🔶	Adapter	LlamaForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/swap-uniba/LLaMAntino-2-7b-hf-ITA"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">swap-uniba/LLaMAntino-2-7b-hf-ITA</a>"	45.77323088190624	Llama 2 Community License	7.0		0	True	🙈	👁	50.55555555555556	53.96398030216369	33.532423208191126	35.0	24.295302013422816	45.45927084511112	48.916213374427244	63.034868799504395	100.0	49.374306621370714	47.508286764686886	0.12030506441959186	24.681836570629475	24.100219820665426	69.11938518934485	40.482970862913184	68.0
+0	sapienzanlp_Minerva-3B-base-v1.0_bfloat16	bfloat16	pretrained	🟢	Original	MistralForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/sapienzanlp/Minerva-3B-base-v1.0"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">sapienzanlp/Minerva-3B-base-v1.0</a>"	41.83701792171881	Apache 2.0	3.0		0	True	🙈	👁	49.23359098650596	52.79773093447906	30.97269624573379	24.333333333333336	23.221476510067113	48.934170047390545	45.62087699981554	48.50152328821496	100.0	45.47176216254846	46.937293275884066	-0.033345460872866474	22.064438703049753	23.965207913141235	43.23710703078177	37.371442699147025	68.60000000000001
+16	swap-uniba_LLaMAntino-2-chat-7b-hf-ITA_bfloat16	bfloat16	fine-tuned	🔶	Adapter	LlamaForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/swap-uniba/LLaMAntino-2-chat-7b-hf-ITA"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">swap-uniba/LLaMAntino-2-chat-7b-hf-ITA</a>"	39.41202334600491	Llama 2 Community License	7.0		0	True	🙈	👁	46.2034115499462	45.34461746324341	29.266211604095567	28.111111111111107	23.758389261744966	42.88181951386289	42.392736217028414	46.58756852047553	100.0	41.699322128331325	45.99082660952828	0.006128977707520721	8.108838055814852	9.097213196911147	58.875305188040464	39.880897484241906	61.8
+2	sapienzanlp_Minerva-1B-base-v1.0_bfloat16	bfloat16	pretrained	🟢	Original	MistralForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/sapienzanlp/Minerva-1B-base-v1.0"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">sapienzanlp/Minerva-1B-base-v1.0</a>"	38.91315019063004	Apache 2.0	1.0		0	True	🙈	👁	48.12085869829324	54.850361197110416	24.573378839590443	22.666666666666664	26.44295302013423	49.56106111987823	46.22580429357212	49.08730795600027	100.0	45.20836949340911	47.013888888888886	0.040313621284920456	14.386315956732856	16.24451875278343	17.353822380105154	39.74793235626088	60.0
+1	sapienzanlp_Minerva-350M-base-v1.0_bfloat16	bfloat16	pretrained	🟢	Original	MistralForCausalLM	"<a target=""_blank"" href=""https://huggingface.co/sapienzanlp/Minerva-350M-base-v1.0"" style=""color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;"">sapienzanlp/Minerva-350M-base-v1.0</a>"	37.29084548916654	Apache 2.0	0.35		0	True	🙈	👁	45.17543859649123	37.91598801552352	24.40273037542662	22.88888888888889	53.8255033557047	42.03399318323408	40.00324919625145	46.79714365710485	100.0	38.049817139468225	44.255424938736375	-0.01382899490742639	10.341357559414417	22.94165519039672	4.978320972441255	43.74869124165633	56.599999999999994

src/populate.py

@@ -15,6 +15,10 @@ def get_leaderboard_df(results_path: str, requests_path: str, cols: list, benchm
 
     df = pd.DataFrame.from_records(all_data_json)
     df = df.sort_values(by=[AutoEvalColumn.average.name], ascending=False)
+
+    if df.shape[0]:
+        df.to_csv("latest_results.tsv", sep="\t")
+
     df = df[cols].round(decimals=2)
 
     # filter out if any of the benchmarks have not been produced