Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ 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
+1460000/default/checkpoint filter=lfs diff=lfs merge=lfs -text

1460000/_CHECKPOINT_METADATA

@@ -0,0 +1 @@
+{"init_timestamp_nsecs": 1720050926314731899, "commit_timestamp_nsecs": 1720051034869877474}

1460000/commit_success.txt

@@ -0,0 +1 @@
+Checkpoint commit was successful to gs://craiyon_models_europe_west4/cappa_datacomp/1460000

1460000/default/_METADATA

@@ -0,0 +1 @@
+{"tree_metadata": {"('params', 'logit_bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "logit_bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'logit_scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "logit_scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'embeddings', 'embeddings', 'embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "embeddings", "key_type": 2}, {"key": "embeddings", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'attention', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attention", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'attention', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attention", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'attention', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attention", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'attention', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attention", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'cross_attention', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "cross_attention", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'cross_attention', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "cross_attention", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'cross_attention', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "cross_attention", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'cross_attention', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "cross_attention", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'mlp', 'wi_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "wi_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'mlp', 'wi_1', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "wi_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'mlp', 'wo', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "wo", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'post_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "post_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'encoder', 'layers', 'post_cross_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "post_cross_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'logits', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "logits", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'text', 'logits', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "text", "key_type": 2}, {"key": "logits", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'embeddings', 'patch_embeds', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "embeddings", "key_type": 2}, {"key": "patch_embeds", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'embeddings', 'position_embeds', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "embeddings", "key_type": 2}, {"key": "position_embeds", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'embeddings', 'registers', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "embeddings", "key_type": 2}, {"key": "registers", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'encoder', 'layers', 'attention', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attention", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'encoder', 'layers', 'attention', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attention", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'encoder', 'layers', 'attention', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attention", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'encoder', 'layers', 'attention', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attention", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'encoder', 'layers', 'mlp', 'wi_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "wi_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'encoder', 'layers', 'mlp', 'wi_1', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "wi_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'encoder', 'layers', 'mlp', 'wo', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "wo", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}, "('params', 'vision', 'encoder', 'layers', 'post_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "vision", "key_type": 2}, {"key": "encoder", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "post_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": true}}}, "use_zarr3": false}

1460000/default/checkpoint

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ab0d575578ee4a8940f80e51571735979ead4005315ea772779c412d30b33b6
+size 2705819122

1460000/default/commit_success.txt

@@ -0,0 +1 @@
+Checkpoint commit was successful to gs://craiyon_models_europe_west4/cappa_datacomp/1460000/default

1460000/metrics/commit_success.txt

@@ -0,0 +1 @@
+Checkpoint commit was successful to gs://craiyon_models_europe_west4/cappa_datacomp/1460000/metrics

config.json

@@ -0,0 +1,70 @@
+{
+  "text_config": {
+    "float32_logits": true,
+    "dtype": "bfloat16",
+    "activations": [
+      "gelu",
+      "linear"
+    ],
+    "normalize_qk": false,
+    "use_bias": false,
+    "force_scale": false,
+    "attention_dropout": 0.0,
+    "mlp_dropout_rate": 0.0,
+    "unroll": 100,
+    "remat_policy": "none",
+    "eos_token_id": 2,
+    "mask_token_id": 4,
+    "pad_token_id": 3,
+    "bos_token_id": 1,
+    "masked_pred_prob": 0.75,
+    "is_decoder": true,
+    "pool_type": null,
+    "num_queries": 1,
+    "vocab_size": 65536,
+    "hidden_size": 1024,
+    "max_length": 64,
+    "num_layers": 12,
+    "use_rmsnorm": true,
+    "ln_type": "normformer",
+    "num_heads": 16,
+    "position_embedding_type": "rotary",
+    "use_causal_mask": true,
+    "mlp_dim": 3072
+  },
+  "vision_config": {
+    "float32_logits": true,
+    "position_embedding_type": "learnt",
+    "position_embedding_shape": null,
+    "position_embedding_factorized": false,
+    "dtype": "bfloat16",
+    "activations": [
+      "gelu",
+      "linear"
+    ],
+    "normalize_qk": false,
+    "use_bias": false,
+    "force_scale": false,
+    "attention_dropout": 0.0,
+    "mlp_dropout_rate": 0.0,
+    "pool_type": null,
+    "unroll": 100,
+    "registers": 8,
+    "keep_registers": true,
+    "remat_policy": "none",
+    "num_queries": 1,
+    "image_size": 256,
+    "hidden_size": 1024,
+    "patch_size": 16,
+    "num_layers": 24,
+    "use_rmsnorm": true,
+    "ln_type": "normformer",
+    "num_heads": 16,
+    "use_causal_mask": false,
+    "mlp_dim": 3072
+  },
+  "projection_dim": 1024,
+  "logit_scale_init_value": 2.3,
+  "logit_bias_init_value": -10.0,
+  "dtype": "bfloat16"
+}

special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,52 @@
+{
+  "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": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "[UNK]"
+}