Training in progress, step 48000, checkpoint

checkpoint-48000/config.json

@@ -0,0 +1,32 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 5,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 4,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 512,
+  "n_embd": 256,
+  "n_head": 8,
+  "n_inner": 1024,
+  "n_layer": 8,
+  "n_positions": 512,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.2",
+  "use_cache": true,
+  "vocab_size": 8192
+}

checkpoint-48000/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 5,
+  "eos_token_id": 4,
+  "transformers_version": "4.45.2"
+}

checkpoint-48000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ccc48c8c65731a804067f556f62c072fa9de3b4f68203d34febbcf1fa73162a
+size 34197144

checkpoint-48000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd0292f3c2853048b22541d18e795ee884eeaa8765f02903b8182892e42e45f9
+size 68458234

checkpoint-48000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf578e8f234e25aba6dcacfdb2a91ed6ad1a96c1a09c62431ca5ecaad5deb62a
+size 14308

checkpoint-48000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21e5243c185c59be5f25f6ee6c277ab9cdefba0458edd2ccad2af317520f936d
+size 1000

checkpoint-48000/special_tokens_map.json

@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "[BOS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "[EOS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "[EOS]"
+}

checkpoint-48000/tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[EOS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "[BOS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "[BOS]",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "[EOS]",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[EOS]",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}

checkpoint-48000/trainer_state.json

@@ -0,0 +1,222 @@
+{
+  "best_metric": 3.916428565979004,
+  "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_recent/fr/wikipedia_clm_13_full/checkpoint-44000",
+  "epoch": 23.278370514064015,
+  "eval_steps": 4000,
+  "global_step": 48000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.9398642095053347,
+      "grad_norm": 1.9506170749664307,
+      "learning_rate": 4e-05,
+      "loss": 6.8417,
+      "step": 4000
+    },
+    {
+      "epoch": 1.9398642095053347,
+      "eval_loss": 5.885306358337402,
+      "eval_runtime": 1.7357,
+      "eval_samples_per_second": 984.627,
+      "eval_steps_per_second": 61.647,
+      "step": 4000
+    },
+    {
+      "epoch": 3.8797284190106693,
+      "grad_norm": 2.0695996284484863,
+      "learning_rate": 7.999000000000001e-05,
+      "loss": 5.4687,
+      "step": 8000
+    },
+    {
+      "epoch": 3.8797284190106693,
+      "eval_loss": 5.096133232116699,
+      "eval_runtime": 1.7474,
+      "eval_samples_per_second": 978.005,
+      "eval_steps_per_second": 61.233,
+      "step": 8000
+    },
+    {
+      "epoch": 5.819592628516004,
+      "grad_norm": 2.068264961242676,
+      "learning_rate": 0.00011996999999999999,
+      "loss": 4.7775,
+      "step": 12000
+    },
+    {
+      "epoch": 5.819592628516004,
+      "eval_loss": 4.6049275398254395,
+      "eval_runtime": 1.7509,
+      "eval_samples_per_second": 976.075,
+      "eval_steps_per_second": 61.112,
+      "step": 12000
+    },
+    {
+      "epoch": 7.759456838021339,
+      "grad_norm": 2.1271705627441406,
+      "learning_rate": 0.00015995,
+      "loss": 4.3147,
+      "step": 16000
+    },
+    {
+      "epoch": 7.759456838021339,
+      "eval_loss": 4.338662147521973,
+      "eval_runtime": 1.7366,
+      "eval_samples_per_second": 984.104,
+      "eval_steps_per_second": 61.614,
+      "step": 16000
+    },
+    {
+      "epoch": 9.699321047526674,
+      "grad_norm": 2.0099985599517822,
+      "learning_rate": 0.00019993,
+      "loss": 4.0005,
+      "step": 20000
+    },
+    {
+      "epoch": 9.699321047526674,
+      "eval_loss": 4.175787925720215,
+      "eval_runtime": 1.7431,
+      "eval_samples_per_second": 980.43,
+      "eval_steps_per_second": 61.384,
+      "step": 20000
+    },
+    {
+      "epoch": 11.639185257032008,
+      "grad_norm": 2.030918598175049,
+      "learning_rate": 0.0002399,
+      "loss": 3.7631,
+      "step": 24000
+    },
+    {
+      "epoch": 11.639185257032008,
+      "eval_loss": 4.070191860198975,
+      "eval_runtime": 1.7392,
+      "eval_samples_per_second": 982.641,
+      "eval_steps_per_second": 61.523,
+      "step": 24000
+    },
+    {
+      "epoch": 13.579049466537342,
+      "grad_norm": 1.9401448965072632,
+      "learning_rate": 0.00027989000000000003,
+      "loss": 3.5859,
+      "step": 28000
+    },
+    {
+      "epoch": 13.579049466537342,
+      "eval_loss": 4.008608341217041,
+      "eval_runtime": 1.7384,
+      "eval_samples_per_second": 983.069,
+      "eval_steps_per_second": 61.55,
+      "step": 28000
+    },
+    {
+      "epoch": 15.518913676042677,
+      "grad_norm": 1.8577344417572021,
+      "learning_rate": 0.00031987,
+      "loss": 3.4474,
+      "step": 32000
+    },
+    {
+      "epoch": 15.518913676042677,
+      "eval_loss": 3.984750986099243,
+      "eval_runtime": 1.7518,
+      "eval_samples_per_second": 975.566,
+      "eval_steps_per_second": 61.08,
+      "step": 32000
+    },
+    {
+      "epoch": 17.458777885548013,
+      "grad_norm": 1.8485515117645264,
+      "learning_rate": 0.00035985,
+      "loss": 3.3504,
+      "step": 36000
+    },
+    {
+      "epoch": 17.458777885548013,
+      "eval_loss": 3.9548492431640625,
+      "eval_runtime": 1.1701,
+      "eval_samples_per_second": 1460.507,
+      "eval_steps_per_second": 91.442,
+      "step": 36000
+    },
+    {
+      "epoch": 19.398642095053347,
+      "grad_norm": 1.7133734226226807,
+      "learning_rate": 0.00039983000000000005,
+      "loss": 3.2673,
+      "step": 40000
+    },
+    {
+      "epoch": 19.398642095053347,
+      "eval_loss": 3.9505419731140137,
+      "eval_runtime": 1.7615,
+      "eval_samples_per_second": 970.203,
+      "eval_steps_per_second": 60.744,
+      "step": 40000
+    },
+    {
+      "epoch": 21.33850630455868,
+      "grad_norm": 1.6587821245193481,
+      "learning_rate": 0.00043981,
+      "loss": 3.1985,
+      "step": 44000
+    },
+    {
+      "epoch": 21.33850630455868,
+      "eval_loss": 3.916428565979004,
+      "eval_runtime": 1.7525,
+      "eval_samples_per_second": 975.17,
+      "eval_steps_per_second": 61.055,
+      "step": 44000
+    },
+    {
+      "epoch": 23.278370514064015,
+      "grad_norm": 1.5722904205322266,
+      "learning_rate": 0.00047979,
+      "loss": 3.1395,
+      "step": 48000
+    },
+    {
+      "epoch": 23.278370514064015,
+      "eval_loss": 3.9289326667785645,
+      "eval_runtime": 1.7436,
+      "eval_samples_per_second": 980.132,
+      "eval_steps_per_second": 61.366,
+      "step": 48000
+    }
+  ],
+  "logging_steps": 4000,
+  "max_steps": 400000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 194,
+  "save_steps": 4000,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 3,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 1
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3725406256889856.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-48000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d9c6412c54f54331c78b98dd2478cc1f092d5793e24f77d250ec7eca3e8c5bb
+size 5368