ninagroot/Llama-360Mtest

README.md

@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 5.8085
+- Loss: 7.2089
 
 ## Model description
 
@@ -33,9 +33,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0003
-- train_batch_size: 128
+- train_batch_size: 32
 - eval_batch_size: 8
 - seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 100
@@ -46,26 +48,18 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| No log        | 1.0   | 2    | 9.5793          |
-| No log        | 2.0   | 4    | 9.4152          |
-| No log        | 3.0   | 6    | 9.1316          |
-| No log        | 4.0   | 8    | 8.7380          |
-| No log        | 5.0   | 10   | 8.3188          |
-| No log        | 6.0   | 12   | 7.9744          |
-| No log        | 7.0   | 14   | 7.7319          |
-| No log        | 8.0   | 16   | 7.5281          |
-| No log        | 9.0   | 18   | 7.3462          |
-| 7.7273        | 10.0  | 20   | 7.1425          |
-| 7.7273        | 11.0  | 22   | 6.9576          |
-| 7.7273        | 12.0  | 24   | 6.8328          |
-| 7.7273        | 13.0  | 26   | 6.7287          |
-| 7.7273        | 14.0  | 28   | 6.5925          |
-| 7.7273        | 15.0  | 30   | 6.4616          |
-| 7.7273        | 16.0  | 32   | 6.3364          |
-| 7.7273        | 17.0  | 34   | 6.1597          |
-| 7.7273        | 18.0  | 36   | 6.0180          |
-| 7.7273        | 19.0  | 38   | 5.9374          |
-| 4.5836        | 20.0  | 40   | 5.8085          |
+| No log        | 0.57  | 1    | 9.6356          |
+| No log        | 1.71  | 3    | 9.5387          |
+| No log        | 2.86  | 5    | 9.3202          |
+| No log        | 4.0   | 7    | 8.9818          |
+| No log        | 4.57  | 8    | 8.7723          |
+| No log        | 5.71  | 10   | 8.3589          |
+| No log        | 6.86  | 12   | 8.0324          |
+| No log        | 8.0   | 14   | 7.7929          |
+| No log        | 8.57  | 15   | 7.7018          |
+| No log        | 9.71  | 17   | 7.5004          |
+| No log        | 10.86 | 19   | 7.3090          |
+| 7.7857        | 11.43 | 20   | 7.2089          |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e00a6cf34998dc1bdc650835cfdca081a8bb9a1394b01e2d5a5f850accc454b7
+oid sha256:6290bb0785659e04767fe8cdf8c82d1f91f177a264ace29bedf516884d83c4de
 size 1408774432

runs/Apr16_13-14-33_gcn71.local.snellius.surf.nl/events.out.tfevents.1713266086.gcn71.local.snellius.surf.nl.1705328.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57b605453c199907b3890431210df38df7c4795497d764ff6c0ed33fda10b1e9
+size 8311

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dda6674de0b9d1258562617d75116fc36d0786efe81f0ce735857249588d736f
+oid sha256:9642187a459ad7b52158c5702400b6f550ccbc6bfeaae638de778b52a4535a12
 size 4984