Upload VJEPA2ForVideoClassification

README.md

@@ -0,0 +1,199 @@
+---
+library_name: transformers
+tags: []
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]

config.json

@@ -0,0 +1,394 @@
+{
+  "architectures": [
+    "VJEPA2ForVideoClassification"
+  ],
+  "attention_dropout": 0.0,
+  "attention_probs_dropout_prob": 0.0,
+  "crop_size": 256,
+  "drop_path_rate": 0.0,
+  "frames_per_clip": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "Approaching [something] with your camera",
+    "1": "Attaching [something] to [something]",
+    "2": "Bending [something] so that it deforms",
+    "3": "Bending [something] until it breaks",
+    "4": "Burying [something] in [something]",
+    "5": "Closing [something]",
+    "6": "Covering [something] with [something]",
+    "7": "Digging [something] out of [something]",
+    "8": "Dropping [something] behind [something]",
+    "9": "Dropping [something] in front of [something]",
+    "10": "Dropping [something] into [something]",
+    "11": "Dropping [something] next to [something]",
+    "12": "Dropping [something] onto [something]",
+    "13": "Failing to put [something] into [something] because [something] does not fit",
+    "14": "Folding [something]",
+    "15": "Hitting [something] with [something]",
+    "16": "Holding [something]",
+    "17": "Holding [something] behind [something]",
+    "18": "Holding [something] in front of [something]",
+    "19": "Holding [something] next to [something]",
+    "20": "Holding [something] over [something]",
+    "21": "Laying [something] on the table on its side, not upright",
+    "22": "Letting [something] roll along a flat surface",
+    "23": "Letting [something] roll down a slanted surface",
+    "24": "Letting [something] roll up a slanted surface, so it rolls back down",
+    "25": "Lifting a surface with [something] on it but not enough for it to slide down",
+    "26": "Lifting a surface with [something] on it until it starts sliding down",
+    "27": "Lifting [something] up completely without letting it drop down",
+    "28": "Lifting [something] up completely, then letting it drop down",
+    "29": "Lifting [something] with [something] on it",
+    "30": "Lifting up one end of [something] without letting it drop down",
+    "31": "Lifting up one end of [something], then letting it drop down",
+    "32": "Moving away from [something] with your camera",
+    "33": "Moving [part] of [something]",
+    "34": "Moving [something] across a surface until it falls down",
+    "35": "Moving [something] across a surface without it falling down",
+    "36": "Moving [something] and [something] away from each other",
+    "37": "Moving [something] and [something] closer to each other",
+    "38": "Moving [something] and [something] so they collide with each other",
+    "39": "Moving [something] and [something] so they pass each other",
+    "40": "Moving [something] away from [something]",
+    "41": "Moving [something] away from the camera",
+    "42": "Moving [something] closer to [something]",
+    "43": "Moving [something] down",
+    "44": "Moving [something] towards the camera",
+    "45": "Moving [something] up",
+    "46": "Opening [something]",
+    "47": "Picking [something] up",
+    "48": "Piling [something] up",
+    "49": "Plugging [something] into [something]",
+    "50": "Plugging [something] into [something] but pulling it right out as you remove your hand",
+    "51": "Poking a hole into [some substance]",
+    "52": "Poking a hole into [something soft]",
+    "53": "Poking a stack of [something] so the stack collapses",
+    "54": "Poking a stack of [something] without the stack collapsing",
+    "55": "Poking [something] so it slightly moves",
+    "56": "Poking [something] so lightly that it doesn't or almost doesn't move",
+    "57": "Poking [something] so that it falls over",
+    "58": "Poking [something] so that it spins around",
+    "59": "Pouring [something] into [something]",
+    "60": "Pouring [something] into [something] until it overflows",
+    "61": "Pouring [something] onto [something]",
+    "62": "Pouring [something] out of [something]",
+    "63": "Pretending or failing to wipe [something] off of [something]",
+    "64": "Pretending or trying and failing to twist [something]",
+    "65": "Pretending to be tearing [something that is not tearable]",
+    "66": "Pretending to close [something] without actually closing it",
+    "67": "Pretending to open [something] without actually opening it",
+    "68": "Pretending to pick [something] up",
+    "69": "Pretending to poke [something]",
+    "70": "Pretending to pour [something] out of [something], but [something] is empty",
+    "71": "Pretending to put [something] behind [something]",
+    "72": "Pretending to put [something] into [something]",
+    "73": "Pretending to put [something] next to [something]",
+    "74": "Pretending to put [something] on a surface",
+    "75": "Pretending to put [something] onto [something]",
+    "76": "Pretending to put [something] underneath [something]",
+    "77": "Pretending to scoop [something] up with [something]",
+    "78": "Pretending to spread air onto [something]",
+    "79": "Pretending to sprinkle air onto [something]",
+    "80": "Pretending to squeeze [something]",
+    "81": "Pretending to take [something] from [somewhere]",
+    "82": "Pretending to take [something] out of [something]",
+    "83": "Pretending to throw [something]",
+    "84": "Pretending to turn [something] upside down",
+    "85": "Pulling [something] from behind of [something]",
+    "86": "Pulling [something] from left to right",
+    "87": "Pulling [something] from right to left",
+    "88": "Pulling [something] onto [something]",
+    "89": "Pulling [something] out of [something]",
+    "90": "Pulling two ends of [something] but nothing happens",
+    "91": "Pulling two ends of [something] so that it gets stretched",
+    "92": "Pulling two ends of [something] so that it separates into two pieces",
+    "93": "Pushing [something] from left to right",
+    "94": "Pushing [something] from right to left",
+    "95": "Pushing [something] off of [something]",
+    "96": "Pushing [something] onto [something]",
+    "97": "Pushing [something] so it spins",
+    "98": "Pushing [something] so that it almost falls off but doesn't",
+    "99": "Pushing [something] so that it falls off the table",
+    "100": "Pushing [something] so that it slightly moves",
+    "101": "Pushing [something] with [something]",
+    "102": "Putting [number of] [something] onto [something]",
+    "103": "Putting [something] and [something] on the table",
+    "104": "Putting [something] behind [something]",
+    "105": "Putting [something] in front of [something]",
+    "106": "Putting [something] into [something]",
+    "107": "Putting [something] next to [something]",
+    "108": "Putting [something] on a flat surface without letting it roll",
+    "109": "Putting [something] on a surface",
+    "110": "Putting [something] on the edge of [something] so it is not supported and falls down",
+    "111": "Putting [something] onto a slanted surface but it doesn't glide down",
+    "112": "Putting [something] onto [something]",
+    "113": "Putting [something] onto [something else that cannot support it] so it falls down",
+    "114": "Putting [something similar to other things that are already on the table]",
+    "115": "Putting [something] that can't roll onto a slanted surface, so it slides down",
+    "116": "Putting [something] that can't roll onto a slanted surface, so it stays where it is",
+    "117": "Putting [something that cannot actually stand upright] upright on the table, so it falls on its side",
+    "118": "Putting [something] underneath [something]",
+    "119": "Putting [something] upright on the table",
+    "120": "Putting [something], [something] and [something] on the table",
+    "121": "Removing [something], revealing [something] behind",
+    "122": "Rolling [something] on a flat surface",
+    "123": "Scooping [something] up with [something]",
+    "124": "Showing a photo of [something] to the camera",
+    "125": "Showing [something] behind [something]",
+    "126": "Showing [something] next to [something]",
+    "127": "Showing [something] on top of [something]",
+    "128": "Showing [something] to the camera",
+    "129": "Showing that [something] is empty",
+    "130": "Showing that [something] is inside [something]",
+    "131": "[Something] being deflected from [something]",
+    "132": "[Something] colliding with [something] and both are being deflected",
+    "133": "[Something] colliding with [something] and both come to a halt",
+    "134": "[Something] falling like a feather or paper",
+    "135": "[Something] falling like a rock",
+    "136": "Spilling [something] behind [something]",
+    "137": "Spilling [something] next to [something]",
+    "138": "Spilling [something] onto [something]",
+    "139": "Spinning [something] so it continues spinning",
+    "140": "Spinning [something] that quickly stops spinning",
+    "141": "Spreading [something] onto [something]",
+    "142": "Sprinkling [something] onto [something]",
+    "143": "Squeezing [something]",
+    "144": "Stacking [number of] [something]",
+    "145": "Stuffing [something] into [something]",
+    "146": "Taking [one of many similar things on the table]",
+    "147": "Taking [something] from [somewhere]",
+    "148": "Taking [something] out of [something]",
+    "149": "Tearing [something] into two pieces",
+    "150": "Tearing [something] just a little bit",
+    "151": "Throwing [something]",
+    "152": "Throwing [something] against [something]",
+    "153": "Throwing [something] in the air and catching it",
+    "154": "Throwing [something] in the air and letting it fall",
+    "155": "Throwing [something] onto a surface",
+    "156": "Tilting [something] with [something] on it slightly so it doesn't fall down",
+    "157": "Tilting [something] with [something] on it until it falls off",
+    "158": "Tipping [something] over",
+    "159": "Tipping [something] with [something in it] over, so [something in it] falls out",
+    "160": "Touching (without moving) [part] of [something]",
+    "161": "Trying but failing to attach [something] to [something] because it doesn't stick",
+    "162": "Trying to bend [something unbendable] so nothing happens",
+    "163": "Trying to pour [something] into [something], but missing so it spills next to it",
+    "164": "Turning [something] upside down",
+    "165": "Turning the camera downwards while filming [something]",
+    "166": "Turning the camera left while filming [something]",
+    "167": "Turning the camera right while filming [something]",
+    "168": "Turning the camera upwards while filming [something]",
+    "169": "Twisting (wringing) [something] wet until water comes out",
+    "170": "Twisting [something]",
+    "171": "Uncovering [something]",
+    "172": "Unfolding [something]",
+    "173": "Wiping [something] off of [something]",
+    "174": "fall",
+    "175": "before-fall",
+    "176": "after-fall"
+  },
+  "image_size": 256,
+  "in_chans": 3,
+  "initializer_range": 0.02,
+  "label2id": {
+    "Approaching [something] with your camera": 0,
+    "Attaching [something] to [something]": 1,
+    "Bending [something] so that it deforms": 2,
+    "Bending [something] until it breaks": 3,
+    "Burying [something] in [something]": 4,
+    "Closing [something]": 5,
+    "Covering [something] with [something]": 6,
+    "Digging [something] out of [something]": 7,
+    "Dropping [something] behind [something]": 8,
+    "Dropping [something] in front of [something]": 9,
+    "Dropping [something] into [something]": 10,
+    "Dropping [something] next to [something]": 11,
+    "Dropping [something] onto [something]": 12,
+    "Failing to put [something] into [something] because [something] does not fit": 13,
+    "Folding [something]": 14,
+    "Hitting [something] with [something]": 15,
+    "Holding [something]": 16,
+    "Holding [something] behind [something]": 17,
+    "Holding [something] in front of [something]": 18,
+    "Holding [something] next to [something]": 19,
+    "Holding [something] over [something]": 20,
+    "Laying [something] on the table on its side, not upright": 21,
+    "Letting [something] roll along a flat surface": 22,
+    "Letting [something] roll down a slanted surface": 23,
+    "Letting [something] roll up a slanted surface, so it rolls back down": 24,
+    "Lifting [something] up completely without letting it drop down": 27,
+    "Lifting [something] up completely, then letting it drop down": 28,
+    "Lifting [something] with [something] on it": 29,
+    "Lifting a surface with [something] on it but not enough for it to slide down": 25,
+    "Lifting a surface with [something] on it until it starts sliding down": 26,
+    "Lifting up one end of [something] without letting it drop down": 30,
+    "Lifting up one end of [something], then letting it drop down": 31,
+    "Moving [part] of [something]": 33,
+    "Moving [something] across a surface until it falls down": 34,
+    "Moving [something] across a surface without it falling down": 35,
+    "Moving [something] and [something] away from each other": 36,
+    "Moving [something] and [something] closer to each other": 37,
+    "Moving [something] and [something] so they collide with each other": 38,
+    "Moving [something] and [something] so they pass each other": 39,
+    "Moving [something] away from [something]": 40,
+    "Moving [something] away from the camera": 41,
+    "Moving [something] closer to [something]": 42,
+    "Moving [something] down": 43,
+    "Moving [something] towards the camera": 44,
+    "Moving [something] up": 45,
+    "Moving away from [something] with your camera": 32,
+    "Opening [something]": 46,
+    "Picking [something] up": 47,
+    "Piling [something] up": 48,
+    "Plugging [something] into [something]": 49,
+    "Plugging [something] into [something] but pulling it right out as you remove your hand": 50,
+    "Poking [something] so it slightly moves": 55,
+    "Poking [something] so lightly that it doesn't or almost doesn't move": 56,
+    "Poking [something] so that it falls over": 57,
+    "Poking [something] so that it spins around": 58,
+    "Poking a hole into [some substance]": 51,
+    "Poking a hole into [something soft]": 52,
+    "Poking a stack of [something] so the stack collapses": 53,
+    "Poking a stack of [something] without the stack collapsing": 54,
+    "Pouring [something] into [something]": 59,
+    "Pouring [something] into [something] until it overflows": 60,
+    "Pouring [something] onto [something]": 61,
+    "Pouring [something] out of [something]": 62,
+    "Pretending or failing to wipe [something] off of [something]": 63,
+    "Pretending or trying and failing to twist [something]": 64,
+    "Pretending to be tearing [something that is not tearable]": 65,
+    "Pretending to close [something] without actually closing it": 66,
+    "Pretending to open [something] without actually opening it": 67,
+    "Pretending to pick [something] up": 68,
+    "Pretending to poke [something]": 69,
+    "Pretending to pour [something] out of [something], but [something] is empty": 70,
+    "Pretending to put [something] behind [something]": 71,
+    "Pretending to put [something] into [something]": 72,
+    "Pretending to put [something] next to [something]": 73,
+    "Pretending to put [something] on a surface": 74,
+    "Pretending to put [something] onto [something]": 75,
+    "Pretending to put [something] underneath [something]": 76,
+    "Pretending to scoop [something] up with [something]": 77,
+    "Pretending to spread air onto [something]": 78,
+    "Pretending to sprinkle air onto [something]": 79,
+    "Pretending to squeeze [something]": 80,
+    "Pretending to take [something] from [somewhere]": 81,
+    "Pretending to take [something] out of [something]": 82,
+    "Pretending to throw [something]": 83,
+    "Pretending to turn [something] upside down": 84,
+    "Pulling [something] from behind of [something]": 85,
+    "Pulling [something] from left to right": 86,
+    "Pulling [something] from right to left": 87,
+    "Pulling [something] onto [something]": 88,
+    "Pulling [something] out of [something]": 89,
+    "Pulling two ends of [something] but nothing happens": 90,
+    "Pulling two ends of [something] so that it gets stretched": 91,
+    "Pulling two ends of [something] so that it separates into two pieces": 92,
+    "Pushing [something] from left to right": 93,
+    "Pushing [something] from right to left": 94,
+    "Pushing [something] off of [something]": 95,
+    "Pushing [something] onto [something]": 96,
+    "Pushing [something] so it spins": 97,
+    "Pushing [something] so that it almost falls off but doesn't": 98,
+    "Pushing [something] so that it falls off the table": 99,
+    "Pushing [something] so that it slightly moves": 100,
+    "Pushing [something] with [something]": 101,
+    "Putting [number of] [something] onto [something]": 102,
+    "Putting [something similar to other things that are already on the table]": 114,
+    "Putting [something that cannot actually stand upright] upright on the table, so it falls on its side": 117,
+    "Putting [something] and [something] on the table": 103,
+    "Putting [something] behind [something]": 104,
+    "Putting [something] in front of [something]": 105,
+    "Putting [something] into [something]": 106,
+    "Putting [something] next to [something]": 107,
+    "Putting [something] on a flat surface without letting it roll": 108,
+    "Putting [something] on a surface": 109,
+    "Putting [something] on the edge of [something] so it is not supported and falls down": 110,
+    "Putting [something] onto [something else that cannot support it] so it falls down": 113,
+    "Putting [something] onto [something]": 112,
+    "Putting [something] onto a slanted surface but it doesn't glide down": 111,
+    "Putting [something] that can't roll onto a slanted surface, so it slides down": 115,
+    "Putting [something] that can't roll onto a slanted surface, so it stays where it is": 116,
+    "Putting [something] underneath [something]": 118,
+    "Putting [something] upright on the table": 119,
+    "Putting [something], [something] and [something] on the table": 120,
+    "Removing [something], revealing [something] behind": 121,
+    "Rolling [something] on a flat surface": 122,
+    "Scooping [something] up with [something]": 123,
+    "Showing [something] behind [something]": 125,
+    "Showing [something] next to [something]": 126,
+    "Showing [something] on top of [something]": 127,
+    "Showing [something] to the camera": 128,
+    "Showing a photo of [something] to the camera": 124,
+    "Showing that [something] is empty": 129,
+    "Showing that [something] is inside [something]": 130,
+    "Spilling [something] behind [something]": 136,
+    "Spilling [something] next to [something]": 137,
+    "Spilling [something] onto [something]": 138,
+    "Spinning [something] so it continues spinning": 139,
+    "Spinning [something] that quickly stops spinning": 140,
+    "Spreading [something] onto [something]": 141,
+    "Sprinkling [something] onto [something]": 142,
+    "Squeezing [something]": 143,
+    "Stacking [number of] [something]": 144,
+    "Stuffing [something] into [something]": 145,
+    "Taking [one of many similar things on the table]": 146,
+    "Taking [something] from [somewhere]": 147,
+    "Taking [something] out of [something]": 148,
+    "Tearing [something] into two pieces": 149,
+    "Tearing [something] just a little bit": 150,
+    "Throwing [something]": 151,
+    "Throwing [something] against [something]": 152,
+    "Throwing [something] in the air and catching it": 153,
+    "Throwing [something] in the air and letting it fall": 154,
+    "Throwing [something] onto a surface": 155,
+    "Tilting [something] with [something] on it slightly so it doesn't fall down": 156,
+    "Tilting [something] with [something] on it until it falls off": 157,
+    "Tipping [something] over": 158,
+    "Tipping [something] with [something in it] over, so [something in it] falls out": 159,
+    "Touching (without moving) [part] of [something]": 160,
+    "Trying but failing to attach [something] to [something] because it doesn't stick": 161,
+    "Trying to bend [something unbendable] so nothing happens": 162,
+    "Trying to pour [something] into [something], but missing so it spills next to it": 163,
+    "Turning [something] upside down": 164,
+    "Turning the camera downwards while filming [something]": 165,
+    "Turning the camera left while filming [something]": 166,
+    "Turning the camera right while filming [something]": 167,
+    "Turning the camera upwards while filming [something]": 168,
+    "Twisting (wringing) [something] wet until water comes out": 169,
+    "Twisting [something]": 170,
+    "Uncovering [something]": 171,
+    "Unfolding [something]": 172,
+    "Wiping [something] off of [something]": 173,
+    "[Something] being deflected from [something]": 131,
+    "[Something] colliding with [something] and both are being deflected": 132,
+    "[Something] colliding with [something] and both come to a halt": 133,
+    "[Something] falling like a feather or paper": 134,
+    "[Something] falling like a rock": 135,
+    "after-fall": 176,
+    "before-fall": 175,
+    "fall": 174
+  },
+  "layer_norm_eps": 1e-06,
+  "mlp_ratio": 4,
+  "model_type": "vjepa2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "num_pooler_layers": 3,
+  "patch_size": 16,
+  "pred_hidden_size": 384,
+  "pred_mlp_ratio": 4.0,
+  "pred_num_attention_heads": 12,
+  "pred_num_hidden_layers": 12,
+  "pred_num_mask_tokens": 10,
+  "pred_zero_init_mask_tokens": true,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.56.0.dev0",
+  "tubelet_size": 2,
+  "use_SiLU": false,
+  "wide_SiLU": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:720160d7a8f7a8d66439e40e5c0640040159769b3668dadea4343d77b15a62f1
+size 1502030892